October 19, 2021
Disease biomarker discovery and fungal metabolites extraction protocol optimization using GC-MS based metabolomics
Chathuri Gamlath Mohottige
Department of Chemistry
Mississippi State University
Tuesday, October 19, 2021
This dissertation focuses on qualitative analysis of metabolite mixtures using HS-SPME coupled GC-MS and TMS derivatization followed by GC-MS analytical platforms. In the first study, we discovered a biomarker combination to diagnose fungal soft tissue disease in sweet potato at an early stage of disease propagation. We used an HS-SPME GC-MS untargeted metabolomics workflow to analyze the VOC associated with Rhizopus stolonifer infected and healthy sweet potatoes in situ (lab) and simulated warehouse environments. A single combination of 4 biomarkers was able to diagnose R. stolonifer fungal soft tissue disease (AUC = 0.980) and the early stage of the fungal soft rot disease (AUC = 0.999). We were able to detect the biomarkers: 1- propanol, ethyl alcohol, ethyl propionate and 3-methyl-3-buten-1-ol during disease progression in a simulated warehouse environment. Therefore, this study shows the feasibility of early diagnosis of fungal soft tissue disease by a real-time screening of volatile profiles of sweet potato in post-harvest storage.
When considering the study of a particular species metabolome, it is crucial to develop a metabolite extraction protocol. Most fungal metabolites extraction protocols are developed using yeast as the model organism. Therefore, in the second study, the performance of the six different metabolite extraction solvents mixtures was tested with the preferred mix being: butanol:methanol:water (2:1:1, v/v at -20°C) which was used as a single solvent mix to extract both polar and relatively non-polar metabolites simultaneously in a single extraction step. The Macrophomina phaseolina (MP) fungal metabolome was investigated using the solvent mix, and two distinct morphological strains of MP fungi were studied.
Finally, fungal mutualism was studied using untargeted metabolomics. Most often mycorrhizal metabolomics workflows are based on analyzing the Arbuscular Mycorrhizae (AM) colonized root metabolome. But here, we used AM hyphal materials to examine the mutualistic symbiotic association of the AM fungi. Harvesting AM extraradical hypha to assess metabolites is often difficult due to inadequate fungal material and soil contamination. Therefore, we used a modified fast method to extract AM fungal hyphae (~ 2min) to provide a snapshot of the AM metabolome when under salinity stress. All untargeted metabolomic studies included chemometric data analysis and specific biomarkers and or metabolites were determined using multivariate statistics or prediction model building and validating.