Transcriptomics and gene regulatory network inference to identify key regulators of Lactuca sativa disease resistance

Plant pests and pathogens are responsible for a large proportion of crop losses. Particularly devastating are generalist necrotrophic fungal pathogens like Botrytis cinerea and Sclerotinia sclerotiorum which affect many economically crucial crops, including lettuce. Due to the environmental and economic implications of pesticide use and the rise of fungicide-resistant strains, there’s an imperative need to develop disease-resistant crop varieties. In this work, several high-throughput transcriptomic datasets are utilised to identify candidate genes which could be manipulated to develop disease-resistant lettuce cultivars. Firstly, I used data which assessed the pathogen susceptibility and transcriptomes of 114 lettuce samples from 27 diverse accessions post-infection with S. sclerotiorum or B. cinerea. This revealed over 5,000 lettuce genes whose expression correlated with S. sclerotiorum resistance across the diversity panel. In addition, two high-resolution time-series datasets of the transcriptomic response to B. cinerea and S. sclerotiorum infection in lettuce leaves, identifying a core set of 4,362 genes which are differentially expressed in the same direction in response to both pathogens. Utilising all four transcriptomic datasets, I inferred a causal gene regulatory network (GRN), highlighting “hub genes”, key transcription factors which are integral to transcriptional reprogramming upon infection. We selected six of these hub genes to validate their in planta defence function. Four of these lettuce hubs altered B. cinerea resistance when constitutively expressed in Arabidopsis or lettuce. Furthermore, the predicted GRN targets genes regulated downstream of a hub gene in planta with higher accuracy than either random-guessing or co-expression modules. This work, therefore, demonstrates a significant advancement in our understanding of defence-induced transcription reprogramming in a crop species. We have been able to successfully predict hub genes and validate their role in the defence response. These results demonstrate that GRNs can be used to identify key regulators of the response to plant stresses in non-model species.