Modulating seed size in oilseed rape

Seeds of oil crops provide an integrated production platform and refinery for high value fatty acids and protein meal. Brassica oilseed rape (OSR) is the primary source of vegetable oil in N.Europe, with huge potential to maximise output and quality for human and livestock nutrition, diversification of novel food and non-food products including biorenewables and first generation biofuels. An important and profitable part of the UK arable rotation, OSR is a recent crop poorly optimised for yield (UK average ~3.3 t/ha) with low harvest index (harvested biomass/total plant biomass) only ~65% of that achievable for wheat. Seed size plays a key role in the yield, quality and profitability of OSR, with ancillary effects on the early stages of crop establishment through its contribution to seedling vigour. Domestication of many crops has been accompanied by an increase in the relative size of the harvested organ, but this is not the case for the seeds of rape since it became adopted as the major oil crop in temperate regions in the past 30 years. However, there appears to be no theoretical developmental limit to making major advances to increase seed size, which should have benefits in terms of surface:volume ratio, processibility and crop establishment. The weedy thale cress species Arabidopsis thaliana has been used by the global plant research community over the past 20 years as a model organism, from which much of our current deep knowledge of the inner workings of plant growth and development has arisen. The complete DNA sequence of the Arabidopsis genome established in 2000 has enabled rapid advances to be made in understanding the role and function of individual genes in plant development. Brassicas are in the same plant family and the closest crop species, so we are able rapidly to make use of much of this fundamental information to benefit a major UK crop. Recent world-leading research at the University of Bath using Arabidopsis has established that there are two major developmental mechanisms by which seed size can be manipulated in this species. These involve firstly the role of endosperm size. Endosperm is the plant equivalent of the placenta / it obtains food from the mother and passes this to the growing embryo. Extending the growing period of the endosperm using various genetic modifications produces a larger endosperm that is better able to obtain food from the mother. The result is a larger embryo. Secondly, increasing the number of cells within the integument layers greatly increases final seed weight. The integuments form a vessel or bag within which the endosperm and embryo develop, rather like the womb of mammals. Our experiments suggest that the embryo and endosperm grow to fill the available space. Providing a larger space by increasing the size of the integuments, again using genetic modifications to development, results in a larger endosperm and embryo, and ultimately a heavier seed. We expect much of the information gained from studying Arabidopsis to be relevant to Brassica crop seeds. However, we need to check that similar mechanisms operate, since Brassica seed can be up to 400 times larger than Arabidopsis seed. In this project we will combine analytical approaches refined in Arabidopsis to study the relative development of seed size in Brassica. We will assign known variation in Brassica seed size to specific developmental mechanisms and will make use of some unique experimental resources to test hypotheses that similar mechanisms operate in the larger Brassica seeds. We will also determine whether the location of genes identified in Arabidopsis as contributing to modulation of seed size are likely to be located at predictable positions on Brassica chromosomes. The information we obtain should greatly assist plant breeders and biologists in developing appropriate screening methods for selecting material from the Brassica genepool with which to modulate seed size in a predictable manner.

Seed growth in the Brassicaceae is controlled and co-ordinated by three main components: endosperm, seed integument and embryo. Currently we know of 2 mechanisms that increase seed size in the reference species Arabidopsis, 1) increased endosperm proliferation during early seed development and 2) increased growth of seed integuments. In order to establish the specific mechanisms that account for variation in seed size in Brassica, we will identify and select a small number of plant lines that represent extremes in the ranges of seed size in our reference mapping and diversity set Brassica populations. Seed from these lines will be subjected to detailed developmental morphological analyses. This will provide our first indicators to the mechanisms existing in Brassica. We have recently verified that the phenomenon of early endosperm proliferation due to paternal excess occurs in Brassica similar to the observations in Arabidopsis. By carrying out reciprocal interploidy crosses (2x X 4x), we will characterise endosperm-led modulation of seed size. Using monosomic addition lines that provide a full set of B. rapa chromosomes and one haploid member B. oleracea chromosomes for each of the 9 C genome chromosomes, we have recently identified a line that produces significantly larger seed. The data from these studies, together with a comparative genomics approach will help attribute QTL for seed size in Brassica to specific candidates and mechanisms that are being characterised in Arabidopsis. This approach will be used primarily to confirm/refute the association of QTL known to have an effect on seed size with specific genes. In parallel, we will introduce a limited number of candidate genes, identified from Arabidopsis, into Brassica plants under selected endosperm specific promoters. The seeds from the transgenic plants will be analysed for increases in seed size and the mechanisms that confer this allowing us to identify genes/loci in Brassica that modulte seed size.