The cereal endosperm is a major source of calories and protein for the world's population and livestock. The vast majority of the wheat crop, however, is milled prior to being used in baking. The overall aim of the project is to improve the processing quality of British wheat by optimising grain shape and size and will involve the work of six UK research groups. Our main objectives are to study the cellular regulation of grain endosperm development; identify and map sources for natural variation in grain shape and size in wheat and barley; manipulate genes with proven roles in endosperm development from the model species Arabidopsis and maize; and determine the effects of different grain shapes on debranning and flour yield.

SCRI's part in the project is to fine map and clone barley globosum-a (glo-a) gene. Both cereals, barley and wheat belong to the Triticea genus. But, barley as opposed to hexaploid wheat, has a diploid genome with a high level of similarity at the nucleotide sequence level enabling more straightforward exploitation of genetic analysis. Therefore, barley can be used as a model for wheat. Identification of the loci and genes that control barley plant interaction with the wheat stem rust fungi is one of such modelling examples.

In this project we proposed to map and clone a barley gene that is known to affect grain shape. Then the cloned barley gene has the potential to be used to manipulate wheat grain shape via a transgenic approach.

The typical phenotype of the plants carrying glo-a allele is reduced length of the lateral and central spikelets and outer glumes. However, the width of these organs is much less affected, and neither is vegetative development - height of the plants carrying mutated glo-a gene is similar to those of the wild type. Mutations in other genes such as Brachytic and Dwarf have similar effect on grain morphometric properties, but in contrast to glo-a, they also have considerable effect on vegetative development - globosum type brachytics and dwarfs usually are significantly shorter.The glo-a phenotype can be clearly observed before anthesis, suggesting that causal genes act during the floral rather than endosperm development.

Cloning strategy

The F2-based mapping population, initially consisting of about 1,000 recombinant gametes will be genotyped and phenotyped (by the quantitative RT-PCR) following the step-wise approach outlined below.

1. Identification of molecular markers linked to the genes of interest. Regions of interest and initial map saturation will be achieved by using small-size populations of homozygous recessive individuals or bulked segregant analysis.

2.Identification of the flanking markers. Markers that delimit the region will next be mapped using the whole population. This step will identify recombinant F2 lines that have crossing-overs happened proximal and distal to the gene.

3. High resolution mapping. Selected recombinant lines will be used to screen remaining markers identified in the previous step. These lines can also be used to map candidate genes identified from other species using various prediction methods. At this point we expect to identify several co-segregating or tightly linked to the phenotype molecular markers.

4. Physical mapping. These markers will be used to identify BAC clones. BAC contig across the gene will be established. Overlapping BAC clones will be used to generate more markers to reduce the contig length. They will be mapped using selected recombinant population. Or, additional F2 recombinants will be identified if resolution of the initial populations is not sufficient.

5. Chromosome walking. Chromosome walk will be initiated if establishment of the BAC contig based on the initial F2 population fails. It will continue till flanking the gene BAC clones are connected. If too many candidate genes are identified, ultra-high density maps will be generated (5,000 -10,000 gametes) to reduce their numbers.

6. Validation of the candidate genes. We expect to identify three to five candidate genes, that will be analysed using range of techniques that we have in our disposal. We will employ in vitro biochemical assays (if available), genetic tests, transgenesis and reverse genetic screens. The actual combinations of these approaches will largely depend on the functional annotations of the candidate genes and respondent genes and their transcriptional behaviour in the barley cultivars, that can be transformed and where structured populations for the reverse genetic screens exist.