Phenomics to improve disease management and resistance breeding

Phenomics are set to transform crop disease and resistance screening by providing high-throughput, accurate tools and techniques that reduce costs.

Plant pathogens reduce crop yields across the world. Two main strategies are used to combat this: breeding for disease-resistant plants, and fungicide application.

Both approaches depend on identifying and measuring disease signs and symptoms. Today, non-invasive sensor technology, part of the domain of plant phenomics, is crucial for this, allowing high-throughput and repeated measurements on living plants.

While phenomic technologies have been used successfully to measure abiotic stresses such as response to water deficit, methods for biotic stresses such as plant diseases are less advanced. The interaction between plants and pathogens causes a variety of signs and symptoms; however, phenomic sensor technologies are rapidly evolving. The tool kit includes drone and platform-mounted sensors such as RGB, multispectral, hyperspectral, fluorescence and thermal sensors.

GRDC is investing in evaluating and improving these sensor technologies for rapid phenotyping and the following are four examples.

In-field high-throughput phenotyping of net blotch in barley

The most damaging pathogens to barley production in Australia are the net blotches: net form (NFNB) and spot form (SFNB).

Efforts to improve genetic resistance or develop fungicide controls for net blotches are critical to a productive future for Australian barley growers. However, progress in pre-breeding and breeding programs designed to improve net blotch resistance is limited by the lack of a method to accurately phenotype the disease in field trials. Visual scoring can be subjective, time-consuming and relies on expert pathologists.

A recent GRDC investment assembled a Net Blotch Consortium with five programs spanning crop protection, genetics and enabling technologies.

Program 4 of the Consortium will deliver a new method for field phenotyping of net blotch in barley that is high-throughput, accurate and low-cost. This program is being led by Associate Professor Bettina Berger, the scientific director of the Plant Accelerator^®, Australian Plant Phenomics Network^®, at the University of Adelaide.

The aim of Program 4, which started in late 2023, is to deploy field phenotyping to increase the pre-breeding and breeding communities’ capacity for evaluating net blotch resistance.

The methods are designed for field-based phenotyping of the net blotches, with data acquisition protocols that can be implemented by pre-breeders and breeders.

Both ground-based imaging and drone-based imaging is being evaluated for the most robust and practical approach. These are suitable for phenotyping net blotch symptoms across key developmental stages (for example, tillering, stem elongation and heading) to enable genetic studies of net blotch resistance.

The methods are environmentally robust, enabling trials to be designed to establish the interaction between genetics and environment (GxE) in net blotch resistance.

They can also be broadly deployed across relevant investments in crop protection, genetic technologies, GRDC-supported National Variety Trials (NVT) and breeding efforts. These methods are specific to detecting net blotch, rather than broadly measuring plant stress symptoms.

Micro-phenomics: digital phenotyping for enhanced disease resistance

Program 3 of the Net Blotch Consortium investment, led Dr Peter Dracatos at La Trobe University, is developing both a macro and a micro-phenomic analysis to determine the mode of action of specific resistance genes either in isolation or in combination.

These methods are being developed to generate laboratory-based insights into diseases. The overarching aim is to broaden the capabilities of the platform to phenotype both necrotrophic and biotrophic cereal pathosystems at seedling and adult plant stages. The team is focusing on both net blotches as their first pathogen subjects.

This work involves the development of a digital phenotyping platform based on a prototype pioneered at the Leibniz Institute of Plant Genetics and Crop Plant Research (IPK) in Germany.

The platform comprises of two components that perform micro-phenomic and macro-phenomic quantification of net blotch responses on barley.

The macro-phenomic component is a customised RGB camera with a robotic crane providing high-throughput capacity while facilitating accurate quantitative disease ratings. These are based on image analysis software optimised for specific pathogen interactions.

The micro-phenotypic component of the platform uses a Zeiss Axioscan sliding microscope to track the fungus in the plant tissue at early time points when symptoms are not visible to the naked eye.

The group at La Trobe University is collaborating with Analytics for the Australian Grains Industry (AAGI) and IPK to develop image analysis software based on machine learning and AI.

These partnerships will lead to extracting the most meaningful data to enhance current genomic approaches to identify new resistances and understand more about available net blotch resistance sources.

Optimising genetic control of wheat rusts through improved phenotyping

Annual savings from the control of stem, leaf and stripe rust diseases of wheat have been estimated at about $1.5 billion – about $1 billion of which comes from genetic resistance achieved through national rust pathogen monitoring and related RD&E.

Crucial to the development of profitable wheat varieties with adequate resistance is greenhouse and field screening of breeding populations so that lines with acceptable levels of rust resistance can be identified and progressed through the breeding pipeline.

This process is complicated by the three wheat rust diseases having different resistance mechanisms, different rust pathogen epidemiologies, and rust pathogen variability – and these change over time due to the emergence of new pathotypes.

Led by Professor Robert Park and coordinated by Dr Karanjeet Sandhu, the GRDC-invested Australian cereal rust control program (ACRCP) Phase 5 project will improve screening methodologies to quickly, accurately and cost-effectively identify susceptibility to current and emerging rust pathotypes in wheat germplasm.

So far, the research team has optimised conditions for both plant growth and development of the three rusts. The program will also explore the potential for image and sensor-based phenotyping to measure the impact of rust symptoms in wheat under controlled conditions and in field.

This is enabling fast-tracked studies of the expression of the two important types of resistance, all stage resistance (ASR) and adult plant resistance (APR), for all three rusts.

The APR screening cycle can now be completed in 35 days and multi-pathotype testing for APR is possible throughout the year, which is difficult under field conditions.

These accelerated and cost-effective methodologies developed for the screening of APR and ASR will help in the speed-breeding of wheat cultivars.

Phenomics to assist blackleg management

Blackleg crown canker disease, caused by the fungal pathogen Leptosphaeria maculans, is responsible for major canola losses. Blackleg disease can be effectively and efficiently managed by using cultivars with host genetic resistance.

However, with major genes susceptible to breakdown, there is an ongoing need to ensure that new resistant varieties are available. Quantitative resistance (QR) is an important and durable component of host genetic resistance. However, like net blotch, breeding programs are limited by a method to objectively, quickly and accurately phenotype the disease in field trials.

A GRDC-invested project being led by Dr Luke Barrett at CSIRO, which began in late 2023, is developing tools to improve throughput and accuracy and also reduce the cost involved in field phenotyping QR to blackleg in canola. Specifically, the project is developing methods to:

1. automatically quantify crown canker severity from image data;

2. detect blackleg infection in field trials using drones; and

3. understand whether resistance to crown canker can be quantified in the field using non-invasive techniques (such as hyperspectral sensors).

The phenotyping tools described above will be developed with the level of throughput and accuracy required to enable improved efficiency in both the pre-breeding (for example, crop protection, genetic technologies, NVT) and breeding industries.