Plant disease phenotyping, the process of detecting, identifying, and quantifying plant disease, is crucial for understanding plant-pathogen interactions. This understanding is essential for breeding disease-resistant crops and reducing reliance of agriculture on pesticides. Traditionally, phenotyping relied on visual assessment by human experts. However, technological advancements are showing promise in improving phenotyping sensitivity, accuracy, and efficiency. In this thesis, researchers explored novel phenotyping methods, focusing specifically on the interaction between lettuce (Lactuca sativa) and the obligate biotrophic oomycete pathogen Bremia lactucae. B. lactucae colonizes lettuce without visible symptoms until sporangiophores emerge from the leaf, forming a white mold typical for downy mildew pathogens. Researchers aimed to develop methods that aid in characterizing and quantifying the early phases of the disease. Such methods would enable more precise phenotyping which is crucial in breeding for downy mildew resistant cultivars.
Researchers explored two methods to enable imaging B. lactucae in vivo at a microscopic scale. Firstly, by developing a protocol for genetic transformation of B. lactucae to generate strains expressing fluorescent proteins for in vivo fluorescence microscopy. Researchers show that Agrobacterium infiltration of downy mildew colonized tissue can achieve at least transient transgene expression in B. lactucae. Secondly, researchers explored a novel imaging technique, dynamic optical coherence tomography (dOCT), for in vivo imaging without labeling with fluorophores. Dynamic OCT utilizes sub-resolution motion to create functional contrast.
The researchers found that the specific dynamic activity inside B. lactucae hyphae can be utilized to create contrast in dOCT images, enabling label-free in vivo imaging of downy mildew colonization inside lettuce tissue. Furthermore, researchers aimed to relieve the bottlenecks of light microscopy of trypan blue stained samples, a traditional method to visualize oomycete hyphae and study infection on a microscopic level. By combining large-field, high-resolution image acquisition with automated image classification based on a convolutional neural network researchers achieved efficient image acquisition and quantification of downy mildew colonization levels in trypan blue stained microscopy images.
To complement these tools for studying downy mildew infection on a microscopic scale, researchers additionally focused on visualizing and quantifying early downy mildew infection on a macroscopic scale. Researchers discovered that lettuce tissue colonized by downy mildew emits increased blue-green fluorescence under UV-A illumination from around 6 dpi before other visible signs or symptoms. This facilitates early detection and quantification of disease severity. Researchers employed UV-fluorescence imaging and an automated image analysis pipeline for phenotyping a recombinant inbred line population to map quantitative trait loci for downy mildew resistance.
Additionally, the researchers explored microscopic, transcriptomic and metabolomic changes in colonized and blue-green fluorescent tissue to elucidate the origin of the fluorescence signal. Researchers identified mesophyll vacuoles as the primary source of downy mildew-induced blue-green fluorescence and noted a downregulation of photosynthesis and an upregulation of the general phenylpropanoid pathway.
More specifically, researchers discovered the accumulation of the caffeoylquinic acids chlorogenic acid and dicaffeoylquinic acid, which likely contribute to the fluorescence signal. Researchers discuss their findings in the broader context of the past, current, and future developments in plant disease phenotyping.
Tonn, S. (2024). Advanced plant disease phenotyping methods to track and quantify lettuce downy mildew. [Doctoral thesis 1 (Research UU / Graduation UU), Universiteit Utrecht]. Utrecht University. https://doi.org/10.33540/2563
Source: Utrecht University