Genetic dissection of Fusarium secondary metabolite production and virulence

F3702 (Gerhard Adam, 2009-2019)

The main goal of this project part is to investigate the role of secondary metabolite biosynthetic genes (predicted in F3705/LAP3714) by gene disruption and virulence testing. The chemical structure of still unknown metabolites should be elucidated (together with F3715). Another goal was to get insights into the mode of action of compounds and to identify their plant targets (achieved for zearalenone, butenolide and culmorin). Next, it was investigated how the plants cope with such metabolites and metabolize it. Candidate genes on the plant side were of interest. A focus was on glutathione-mediated detoxification of trichothecenes and on a newly identified chemotype (F. graminearum producing NX-toxins). In the third phase the role of auxin biosynthesis by Fusarium was investigated. A challenge was the redundancy of genes (e.g. seven amine oxidases involved in auxin biosynthesis had to be inactivated). Therefore, technology development was necessary, e.g. the construction of positive negative selectable fusion genes allowing recycling of transformation markers. The novel tools allowed to generate a septuple knockout mutant which revealed that auxin biosynthesis is indeed a virulence factor of F. graminearum.

 

Regulation of toxin biosynthetic genes and secondary metabolite production by chromatin structure and epigenetics in F. graminearum
F3703 (Joseph STRAUSS, 2009-2019)

We started our engagement into the Fusarium field with the main working hypothesis that chromatin structures and epigenetic mechanisms may regulate secondary metabolism and virulence. This was based on our previous discoveries in other systems and we showed in an initial publication that heterochromatin regulates Fusarium secondary metabolites. This inspired a lot of research into this direction not only in different Fusarium species but also in other ascomycete plant pathogens or endophytes. Our subsequent papers described a novel and conserved chromatin-based regulator of fungal secondary metabolism (KdmB/Kdm5). In extensive collaborations within the consortium (chemical analytics, plant breeding, bioinformatics), we showed that only the actively defending plant is able to elicit the toxin pathways. Based on pathogenicity profiles of a hep1 heterochromatin mutant we suggest that plant metabolites interfere with virulence factor expression at the level of chromatin regulation. We have established fungal chromatin modification analysis during pathogenesis (ip-ChIP) and this will help to unravel pathogenicity-related chromatin processes in situ. Overall, we have contributed a novel concept for the Fusarium-host interaction at the level of chromatin regulation that also opens new possibilities to design fungicides based on specific inhibitors of chromatin factors.

 

Comparative annotation and analysis of multiple Fusarium genomes and the genomes of Brachypodium and crop plants with emphasis on toxin biosynthesis and detoxification reactions

F3705 + LAP3714 (Hans-Werner Mewes, Ulrich Güldener and Klaus FX Mayer, 2009-2019)

The project is focusing on genomics and integration of further -omics data on the cereal plant hosts and the pathogen, the Fusarium graminearum species group and related species. On the pathogen side several genome annotations were achieved or improved with the help of available RNA-Seq data, which aided the experimental partners with gene specific experiments. Beside the general annotation a focus was on secondary metabolite gene clusters. A generic method was developed and successfully applied on multiple genomes. Beside using RNA-Seq data to assemble transcripts to help during annotation the data was also used for quantitative expression analysis (F3703). The genomic resources were further developed to a multi-species resource including integration of -omics data.

The second focus was to make use of plant genome sequences, most notably the 17 Gb hexaploid bread wheat genome that has recently become available. Exploiting these resources allowed addressing the effects of F. graminearum infection. Gene expression data generated by RNA-Seq was mapped against current genome assemblies and annotations and the underlying molecular mechanisms and network-based expression adaptations were studied. By combining the genomic gene expression data with network-based data analyses methods groups of genes that react in a pathogen-responsive or QTL-specific manner were extracted.

 

Metabolomics of plant-Fusarium interactions

F3706 and F3715 (Krska and Schuhmacher, 2009-2019)

The major goals of the subprojects F3706/15 are the establishment of a comprehensive and standardized metabolomics platform for the study plant-Fusarium interactions and its application to investigate fungal small molecule virulence factors and plant defense mechanisms. Initially developed GC-MS methods for profiling of volatile organic compounds and polar, mainly primary metabolites were used. As a major achievement three interdependent cutting-edge metabolomics workflows (global metabolome labelling, tracer-derived labelling and isotope-assisted MS/MS) were developed together with the recently published and freely available MetExtract II software specifically designed to aid in the data evaluation of the respective metabolomics workflows. Recently we have extended our labelling approach by another workflow with the aim to differentiate between Fg and plant metabolites under infection conditions. PhytoLabelboxes have successfully been used for the custom-tailored production of globally 13C enriched T. aestivum and T. durum plants under control and stress conditions (F3708, F3711, Lemmens). The uniformly 13C and 15N-labelled reference samples enabled the reliable detection and study of more than 1500 wheat metabolites after both toxin treatment and pathogen infection. Complementary to this, stable isotope tracers were generated (tryptamine, auxin, butenolide) or purchased (Phe, Trp and Tyr). Tracer-fate studies with 13C labelled exogenous (DON) and endogenous metabolites revealed previously unknown glucose- and glutathione-related plant detoxification products as well as Phe-, Trp- and Tyr-derived wheat submetabolomes, which are currently further evaluated to find resistance related defense compounds (with F3702, F3708, F3711). Time series metabolomics and transcriptomics data are being jointly evaluated together with F3705 and F3711 to investigate metabolic processes from 3 to 96 hours after infection in detail. In a joint effort with F3702 and F3708 we were also able to demonstrate that F. graminearum can hydrolyze the host derived antifungal coumaroyltryptamine and produce auxin from the resulting tryptamine. In cooperation with F3702 we are currently employing fungal H3K27 knock-out strains to elucidate novel infection associated polyketides.

 

Exploring the role and metabolism of zearalenone and other Fusarium compounds in the model plant Arabidopsis

F3707 (Marie-Theres Hauser, 2009-2012)

The aim of this part was to identify responses triggered by zearalenone (ZEN) in the model plant Arabidopsis thaliana. Zearalenone is a strongly estrogenic substance in animals, yet plants lack estrogen receptors. Based on microarray analyses and a genetic screen for ZEN responsive mutants several hypotheses were developed and tested: ZEN interferes with the cell wall integrity response, and the UV-B induced phenylpropanoid biosynthesis pathway. The functional relevance of ZEN induced genes such as an identified UDP-glycosyltransferase (UGT) and an ABC-transporter for the metabolization and secretion of ZEN and its metabolites was investigated. Together with F3706 ZEN metabolization analyses and the quantification of ethylene were done, which revealed that ZEN is inhibiting the biosynthesis of ethylene and that the induced ABC-transporter is involved in the secretion of β-zearalenol and ZEN-4-sulfate. Using the promoters and 3' UTRs of nine strongly and rapidly ZEN inducible genes Arabidopsis reporter lines have been established that allow monitoring the presence of ZEN in planta. Although the project was stopped after the first grant period some results partly supported by this grant could be published in five peer reviewed journals (see part B).

 

Detoxification by plant UDP-glucosyltransferases

F3708 (Gerhard Adam, 2009-2019)

The main goal of this project part is to identify a monocot UDP glucosyltransferase with the capability to inactivate DON to DON-3-O-glucoside. In collaboration with Prof. Gary Muehlbauer group (Univ. of Minnesota), who provided candidates that were specifically DON induced, the gene HvUGT13248 could be validated, which conferred not only resistance to DON but also to nivalenol (NIV) when expressed in yeast. The UGT gene family of Brachypodium consisting of 159 genes was analyzed with support from F3705. A cluster of six Brachypodium genes with high similarity to HvUGT13248 was identified, but only two of them exhibited detoxification capability. We also analyzed orthologous genes from rice and Sorghum bicolor and found that the UGT clusters undergo rapid changes in gene copy number and changes in substrate specificity. In collaboration with the group of Ivan Rayment (University of Wisconsin, Madison) a rice gene similar to HvUGT13248 expressed from a recoded gene in E. coli was purified and could be crystallized. The three-dimensional structure was determined, leading the way to rationally alter substrate specificity determinants. With E. coli expressed affinity purified proteins various 3-O-β-glucosides of trichothecenes could be enzymatically produced and provided to other researchers performing animal feeding trials with these masked mycotoxins. Overexpression of the HvUGT13248 gene in Arabidopsis led to high level DON resistance, without obvious side effects. The gene was introduced also into wheat and conferred resistance against DON and NIV and as well as high level Fusarium spreading resistance. No UGT gene was detected in the QTL region containing the Fhb1 locus. The hypothesis that a UDP-glucose dehydrogenase pseudogene present in the QTL region of susceptible lines affects UGT co-substrate UDP-glucose levels was investigated but could not be substantiated. Furthermore, in this project part also a barley UGT capable of synthesis of ZEN-14- and the novel ZEN-16-glucoside could be identified, and was used to enzymatically produce these glucosides.

 

Functional genomics of Fusarium resistance in wheat

F3711 (Hermann Bürstmayr, 2009-2019)

The detailed analysis of the wheat/Fusarium interaction at the transcriptome level and the identification the causal genes behind major FHB resistance quantitative trait locus (QTL) Fhb1 and Qfhs.ifa-5A are the main tasks of this subproject. The core plant materials for these objectives were two sets of near-isogenic wheat lines with combinations of Fhb1 and Qhfs.ifa-5A in a susceptible or resistant background. RNA sequencing and microarray analysis of Fusarium-challenged wheat heads identified QTL-associated candidate genes and described the underlying biology of the host/pathogen interaction. For gene isolation map-based approaches have been started for both resistance QTL, high resolution maps were developed and a BAC library was generated for the resistant line CM82036 harboring both QTL. A contig spanning the Fhb1 region was established and fine-mapping delimited the QTL region to 860 kb and 28 genes. Eleven of these candidates were excluded as the causal genes using TILLING. A forward genetics approach with DON infiltration for phenotyping identified four mutants showing DON-induced bleaching, FHB spreading and reduced DON detoxification to DON-3-O-glucoside. Sequencing of flow-sorted 3B chromosomes from these mutants and CM82036 is underway and will allow to identfy the underlying common causal factor. Fhb1 enhances FHB resistance also in durum wheat and triticale. We elucidate its effect on mycotoxin content and metabolization in these genetic backgrounds. The identification of genes behind the ‘resistance to initial infection’ QTL Qfhs.ifa-5A is complicated by its position in the low-recombining pericentromeric region of chromosome 5A. Linkage mapping identified two adjacent resistance QTL associated with anther extrusion, suggesting that Qfhs.ifa-5A acts as a passive resistance factor. Deletion mapping increased map resolution significantly and will support fine-mapping and gene isolation.

 

Validation of candidate genes with transgenic plants

F3712 (Eva Stoeger, 2019-2012)

Project part 12 was responsible for the generation of transgenic wheat plants expressing candidate genes related to Fusarium or fungal toxin resistance. The work was conducted in collaboration with the groups of Gerhard Adam (F3702) and Hermann Bürstmayr (F3711), where selection of candidate genes with a putative role in Fusarium resistance was carried out. In line with the original proposal the main activities of our project part started in the second half of the first project phase as most of the candidate genes, as well as growth facilities for transgenic cereals, became available only in years 3 and 4. Biolistic gene transfer into immature embryo-derived callus tissue was carried out using a total of 8 vector constructs for the transfer of the candidate genes. Co-transformation was used to introduce the selectable marker in combination with the various genes of interest and 130 independent transgenic lines were obtained and analysed for the presence and expression of the transgenes. The lines include plants expressing the RPS11 gene or a fungal siderophore esterase (SIE), respectively (provided by F3702). The recombinant proteins were detected in leaves of several of these lines. In collaboration with the Lemmens group preliminary tests to assess resistance to DON were conducted with segregating wheat populations of RPS11 transformants. Further lines were generated with inducible UDP-glucosyltransferases (F3708, F3707) from Brachypodium. The project part was terminated at the end of the first project phase. For most of the transgenic lines T1 seed was available by the end of the project part. The molecular characterization and zygosity determination of the plants was completed, and seeds were passed on to the consortium.

 

Toxicology and metabolism of Fusarium toxins and co-regulated secondary metabolites

F3718 (Doris Marko, 2016-2019)

It is the overall objective of F3718 to support the consortium in a more detailed characterization of toxicological properties and cellular metabolism of the newly identified compounds in mammalian cells to unravel their potential impact on human health. This is done in three distinct work packages: WP1: The toxicological potential of the newly identified type-A trichothecene NX-3 was thoroughly characterized in different mammalian cell systems and compared to that of DON. Besides, other novel metabolites such as the recently discovered DON-sulfates were also thoroughly investigated. Moreover, we screened for glutathione-mediated detoxification of DON in mammalian cells and elucidated the impact of a potential depletion of the glutathione pool in the detoxification of NX-toxins and DON. WP2: Focused on the extension of the stable isotopic labeling (SIL) approach established by partner F3715 for studying metabolism of xenobiotics using tailored bioinformatic evaluation with the MetExtract II software from plant to mammalian cell systems. This allowed for the detailed investigation of human DON and ZEN metabolism. WP3: Combinatory effects between DON and other secondary fungal metabolites, which have been regarded as non-toxic so far such as butenolide or culmorin, were studied in detail. We proved the relevance of these interactions in mammalian cells for the first time.