Breadmaking is one of humankind's oldest technologies, being established some 4,000 years ago. The ability to make leavened bread depends largely on the visco-elastic properties conferred to wheat doughs by the gluten proteins. These allow the entrapment of carbon dioxide released by the yeast, giving rise to a light porous structure. One group of gluten proteins, the high molecular weight (HMW) subunits, are largely responsible for gluten elasticity, and variation in their amount and composition is associated with differences in elasticity (and hence quality) between various types of wheat. These proteins form elastomeric polymers stabilized by inter-chain disulphide bonds, and detailed studies of their structures have led to models for the mechanism of elasticity. This work has also provided a basis for direct improvement of wheat quality by transformation with additional HMW subunit genes.

The high molecular weight (HMW) subunits of wheat glutenin are major determinants of the elastic properties of gluten that allow the use of wheat doughs to make bread, pasta, and a range of other foods. There are both quantitative and qualitative effects of HMW subunits on the quality of the grain, the former being related to differences in the number of expressed HMW subunit genes. We have transformed bread wheat in order to increase the proportions of the HMW subunits and improve the functional properties of the flour. A range of transgene expression levels was obtained with some of the novel subunits present at considerably higher levels than the endogenous subunits. Analysis of T2 seeds expressing transgenes for one or two additional HMW subunits showed stepwise increases in dough elasticity, demonstrating the improvement of the functional properties of wheat by genetic engineering.

SummaryA chimaeric gene was constructed comprising a wheat high molecular weight glutenin subunit gene promoter, a 304-bp sucrose non-fermenting-1-related (SnRK1) protein kinase sequence in the antisense orientation, and the cauli¯ower mosaic virus 35S RNA gene terminator. Transgenic barley plants containing the antisense SnRK1 chimaeric gene were produced by particle bombardment of barley immature embryos with the aim of obtaining plants expressing the antisense SnRK1 sequence in the seeds. Despite the fact that the promoter was expected to be active only in seeds, two independent transgenic lines were found to fail to transmit the transgene to the T 1 generation. These T 0 plants had matured and died before this was discovered, but subsequently four other independent transgenic lines were found to be affected in the same way. Cytological analysis of the pollen grains in these lines showed that about 50% were normal but the rest had arrested at the binucleate stage of development, were small, pear-shaped, contained little or no starch and were non-functional. The presence of antisense SnRK1 transcripts was detected in the anthers of the four lines analyzed and a ubiquitin promoter/UidA (Gus) gene, one of the marker genes codelivered with the antisense gene, was found to be expressed only in the abnormal pollen. Expression analyses con®rmed that SnRK1 is expressed in barley anthers and that expression of one class of SnRK1 transcripts (SnRK1b) was reduced in the abnormal lines. All of the abnormal lines showed approximately 50% seed set, and none of the transgenes were detected in the T 1 generation.