Higher yield, more stress-tolerant crops, thanks to multiplex gene editing
Many genes govern the traits which determine crop yield and tolerance to adverse conditions, and the identification of crops with these desirable traits typically involving screening large cross-bred plant populations. While decades of plant molecular biology has identified many yield and stress tolerance mechanisms, implementation options are still limited. “Although breeding has improved with genetic markers and random mutagenesis(opens in new window), it remains very time- and resource-intensive,” says Dirk Inze(opens in new window), coordinator of the BREEDIT project, which was funded by the European Research Council(opens in new window). According to Inze from the Flemish Institute for Biotechnology, the project host, a major reason is that: “it remains virtually impossible to predict which combination of small effect genes needs to be modified to obtain the desired outcome.” This challenge is compounded by the sheer volume of data involved in trying to represent as many gene combinations as possible from a large collection of genes. Starting with 60 yield-related genes, BREEDIT combined conventional breeding and gene editing techniques to identify gene combinations which result in desired yield or stress tolerance phenotypes. “While we achieved a proof of concept in maize, the BREEDIT approach is applicable to many crops,” adds Inze.
Revolutionising the ‘reverse genetic’ approach
BREEDIT’s novelty was in both starting with a large selection of candidate genes involved in a given crop trait, and developing an approach to test as many higher order gene combinations as possible for improving the trait. Mutations were created by applying CRISPR/Cas9 gene editing(opens in new window). ‘SCRIPT’ constructs, which tell the gene EDITOR where to make the change, were applied to 12 different genes in a maize plant. Cross-breeding these plants with plants containing other SCRIPT constructs – so containing different mutated genes – generated further combinations of mutations. In total, five different SCRIPTs were used, representing 60 genes related to growth. “This approach knocked out more members of gene regulatory networks than experiments only targeting a few genes, resulting in novel gene combinations more profoundly influencing plant traits,” explains Inze. The phenotypes of the plants containing these multiplex gene edits (multiple mutations) were then monitored at different developmental stages. Seedling-stage maize was phenotype screened for interesting growth characteristics, resulting in a dataset of over 6 000 seedlings. These were analysed using advanced machine learning tools. Promising were populations tested at maturity by an automated phenotyping platform, PHENOVISION(opens in new window), using visual and hyperspectral imaging throughout the plants’ life cycle. “Maize plants were found that are able to produce more biomass under both well-watered and drought conditions. Additionally, we identified gene editing that led to modified plant architecture, stem width and improved seed yield,” says Inze.
Commercialisation plans will expand target traits and crops
With Europe facing the negative impacts of climate change on food production, the need for novel, climate-adapted crops becomes ever more urgent. The BREEDIT approach could help speed up the creation of such crops, alongside promoting more sustainable agriculture. Towards this end, the team will next evaluate some of BREEDIT’s most promising edit combinations for further development. Meanwhile, the BREEDIT approach has already attracted attention from the project’s commercial collaborators, resulting in spin-off company RAINBOW CROPS(opens in new window), which will expand the concept to new traits and crops. “BREEDIT’s gene editing success was limited to greenhouses, it now must be replicated in fields. Furthermore, while BREEDIT focused on deactivating genes, we can also experiment with activating genes as knowledge increases about their functions,” notes Inze.
