Genetic Reduction of Energy use and Emissions of Nitrogen through cereal production: GREEN grain

Abstract

• This project initiated the development of a new wheat type for the UK - one with high energy grain, suited to alcohol production and livestock feeding, and with low nitrogen fertiliser requirements hence low environmental impacts. Primary beneficiaries of this 'GREEN grain' are predicted to be wheat growers through reduced costs of crop nutrition, but alcohol processors would also benefit, and emissions of nitrate, ammonia and nitrous oxide would reduce markedly. Wheat seed performance was also improved, but expected benefits for the livestock industry remain to be proven. Findings can be summarised as follows:
• Genetic variation amongst elite and 'global' germplasm indicated potential to breed varieties with the 'GREEN grain' ideotype - having both low canopy nitrogen and low grain protein. Varieties with lower grain protein tended to have higher grain yields when N was withheld.
• Examples of varieties demonstrating desirable traits were Defender (good N capture), Creativ (low leaf N), Solstice (low leaf sheath & stem N), Acropolis (low ear N), Exsept (low straw & chaff N), and Audi (low grain N).
• Canopy and N recovery traits showed genetic variation, but heritabilities were low, variation was often small, and few useful quantitative trait loci (QTL) were discovered, so a wider range of germplasm must now be explored.
• Grain protein had much greater heritability and, with its ease of measurement, breeding for low grain protein appears to offer the greatest scope to reduce N requirements. Genetic variation in grain protein can be as great as the fertiliser effect on grain protein, with fertilised low protein varieties having lower protein than unfertilised high protein varieties.
• Gliadins accounted for over 50% of grain protein and were much the most responsive fraction to nitrogen supply, but there was disappointingly little genetic variation in gliadin:total protein ratio.
• The inverse relationship between protein content and alcohol yield was confirmed, but alcohol yield also showed consistent genetic variation of up to 50 litres per dry tonne unrelated to protein content (probably related to nonstarch polysaccharides).
• Varieties with the best alcohol yields (e.g. Glasgow, Denman, Zebedee, Istabraq) tended to have high grain yields and low grain protein, facilitating combined selection for alcohol yield and reduced N requirements. Maximum alcohol yields were 481 litres per dry tonne, and 4,638 litres per hectare.
• Near infrared spectrometer (NIR) calibrations for alcohol yield and gliadin content were developed that should prove useful for distilling and bioethanol processors, and also for plant breeders.
• Two commercial broiler feeding trials tested diets containing proto-type GREEN grain (6-8% protein instead of normal 10-11% protein). Manure N content and ammonia emissions were reduced, and effects on manure output, bird mortality, health and performance were not significant.
• Analysis of mapping populations returned several potentially useful QTL for protein composition and grain size. These QTL will be validated by Syngenta prior to possible use in their commercial wheat breeding programmes.
• If the industry and the wider environment are to benefit from this work, and from associated research, there is now an urgent need for variety tests by plant breeders, and for National and Recommended Lists, to include criteria determining N requirements.