Integrating microbiology and metabolomics to decode myrtle rust resistance in Rhodamnia rubescens

Abstract

Myrtle rust an emerging forest disease caused by obligate pathogen Austropuccinia psidii infecting myrtaceous plants. This pathogen arrived in Australia in 2010 and since then, it has been confirmed to infect over 350 species of Australian Myrtaceae plants, encompassing 58 different genera. Some species such as Rhodamnia rubescens, are now critically endangered as a result of the disease. This research used a dual approach to investigate possible mechanisms of resistance in R. rubescens which is separated into two chapters. The first chapter aimed to evaluate the fungal communities present in symptomatic and asymptomatic leaves of R. rubescens for antagonistic effect against myrtle rust and observe if microbes play a role in natural resistance in field. Fungal communities were isolated and identified using culture dependent and DNA sequencing of the ITS and 18S gene region. No correlation was observed between the microbes and symptomatic and asymptomatic leaves. From the 141 isolates obtained, 10 isolates from the genera Xylariaceae, Cladosporium, Unknown isolate, Pestalotiopsis, Quambalaria and Neoconiothyrium showed significant potential in suppressing the germination of A. psidii spores in vitro. Volcano analysis of isolates with potential antagonistic ability from in vitro assay provided for the selection of isolates for further investigation in terms of mode of action, metabolite profiles and possible synergistic combinations for enhanced biocontrol efficacy.

The second chapter aimed to compare the metabolite profile to detect differences between the myrtle rust resistant and susceptible phenotypes that could be potential biomarkers to develop screening tools for myrtle rust disease resistance to aid breeding strategies. A multi-omics extraction of leaves from resistant and susceptible plants was performed using methanol and chloroform coupled with ultra-highperformance liquid chromatography and mass spectrometry. This allowed prediction of 53 significant compounds of which eight compounds which had closest match to oxaloacetate, crocin, glycyrrhizic acid, p-coumaric acid, 4-oxo-2-nonenal, threonate, mesaconic acid and corosolic acid were identified as plant metabolites with significant difference between susceptible and resistant phenotype. However, these compounds did not have good discriminative ability to accurately distinguish myrtle rust resistant and susceptible phenotype. Partial least squares-discriminant analysis- variable importance projection analysis combined with univariate and multivariate ROC curve analysis, three unknown compounds (4.05_649.1055m/z, 4.05_627.1232n and 4.05_517.1366m/z) were selected as potential biomarkers with moderate discriminative ability to accurately distinguish the resistant and susceptible phenotype based on the compounds present in the leaves alone.

Through the identification and characterisation of leaf associated fungal isolates and analysis of the metabolomic profile of myrtle rust disease resistant and susceptible phenotypes, this study contributed towards identifying potential biocontrol agents for myrtle rust disease management and biomarkers that could be used in developing screening tools for resistant or susceptible phenotypes in conservation approaches.

Date of Thesis submission	2024
Original language	English
Awarding Institution	Western Sydney University
Supervisor	Michelle Moffitt (Supervisor), Jonathan Plett (Supervisor), Colin Stack (Supervisor) & Oliver Morton (Supervisor)