Search

Fungi from the genus Epichloë form systemic endobiotic infections of cool season grasses, producing a range of host-protective natural products in return for access to nutrients. These infections are asymptomatic during vegetative host growth, with associations between asexual Epichloë spp. and their hosts considered mutualistic. However, the sexual cycle of Epichloë spp. involves virulent growth, characterized by the envelopment and sterilization of a developing host inflorescence by a dense sheath of mycelia known as a stroma. Microscopic analysis of stromata revealed a dramatic increase in hyphal propagation and host degradation compared with asymptomatic tissues. RNAseq was used to identify differentially expressed genes in asymptomatic vs stromatized tissues from 3 diverse Epichloë–host associations. Comparative analysis identified a core set of 135 differentially expressed genes that exhibited conserved transcriptional changes across all 3 associations. The core differentially expressed genes more strongly expressed during virulent growth encode proteins associated with host suppression, digestion, adaptation to the external environment, a biosynthetic gene cluster, and 5 transcription factors that may regulate Epichloë stroma formation. An additional 5 transcription factor encoding differentially expressed genes were suppressed during virulent growth, suggesting they regulate mutualistic processes. Expression of biosynthetic gene clusters for natural products that suppress herbivory was universally suppressed during virulent growth, and additional biosynthetic gene clusters that may encode production of novel host-protective natural products were identified. A comparative analysis of 26 Epichloë genomes found a general decrease in core differentially expressed gene conservation among asexual species, and a specific decrease in conservation for the biosynthetic gene cluster expressed during virulent growth and an unusual uncharacterized gene.

Epichloë endophytes are filamentous fungi (family Clavicipitaceae) that live in symbiotic associations with grasses in the sub family Poöideae. In New Zealand, E. festucae var. lolii confers significant resistance to perennial ryegrass (Lolium perenne) against insect and animal herbivory and is an essential component of pastoral agriculture, where ryegrass is a major forage species. The fungus produces in planta a range of bioactive secondary metabolites, including ergovaline, which has demonstrated bioactivity against the important pasture pest black beetle, but can also cause mammalian toxicosis. We genetically modified E. festucae var. lolii strain AR5 to eliminate key enzymatic steps in the ergovaline pathway to determine if intermediate ergot alkaloid compounds can still provide insecticidal benefits in the absence of the toxic end product ergovaline. Four genes (dmaW, easG, cloA, and lpsB) spanning the pathway were deleted and each deletion mutant was inoculated into five different plant genotypes of perennial ryegrass, which were later harvested for a full chemical analysis of the ergot alkaloid compounds produced. These associations were also used in a black beetle feeding deterrence study. Deterrence was seen with just chanoclavine present, but was cumulative as more intermediate compounds in the pathway were made available. Ergovaline was not detected in any of the deletion associations, indicating that bioactivity towards black beetle can be obtained in the absence of this mammalian toxin.

The association of plants with endosymbiotic micro-organisms poses a particular challenge to metabolomics studies. The presence of endosymbionts can alter metabolic profiles of plant tissues by introducing non-plant metabolites such as fungal specific alkaloids, and by metabolic interactions between the two organisms. An accurate quantification of the endosymbiont and its metabolites is therefore critical for studies of interactions between the two symbionts and the environment.Here, we describe methods that allow the quantification of the ryegrass Neotyphodium lolii fungal endosymbiont and major alkaloids in its host plant Lolium perenne. Fungal concentrations were quantified in total genomic DNA (gDNA) isolated from infected plant tissues by quantitative PCR (qPCR) using primers specific for chitinase A from N. lolii. To quantify the fungal alkaloids, we describe LC-MS based methods which provide coverage of a wide range of alkaloids of the indolediterpene and ergot alkaloid classes, together with peramine.

The identification of quantitative trait loci (QTL) for plant metabolites requires the quantitation of these metabolites across a large range of progeny. We developed a rapid metabolic profiling method using both untargeted and targeted direct infusion tandem mass spectrometry (DIMSMS) with a linear ion trap mass spectrometer yielding sufficient precision and accuracy for the quantification of a large number of metabolites in a high-throughput environment. The untargeted DIMSMS method uses top-down data-dependent fragmentation yielding MS2 and MS3 spectra. We have developed software tools to assess the structural homogeneity of the MS2 and MS3 spectra hence their utility for phenotyping and genetical metabolomics. In addition we used a targeted DIMS(MS) method for rapid quantitation of specific compounds. This method was compared with targeted LC/MS/MS methods for these compounds. The DIMSMS methods showed sufficient precision and accuracy for QTL discovery. We phenotyped 200 individual Lolium perenne genotypes from a mapping population harvested in two consecutive years. Computational and statistical analyses identified 246 nominal m/z bins with sufficient precision and homogeneity for QTL discovery. Comparison of the data for specific metabolites obtained by DIMSMS with the results from targeted LC/MS/MS analysis showed that quantitation by this metabolic profiling method is reasonably accurate. Of the top 100 MS1 bins, 22 ions gave one or more reproducible QTL across the 2 years. Copyright © 2009 John Wiley & Sons, Ltd.