Use of Seed Storage Proteins for Protein Quality Improvements in Nonseed Crops

Besides seeds, a variety of other plant organs are valuable sources of protein. Potato tubers are the most important noncereal food crop, since they are consumed by humans and animals and used in the manufacture of starch and alcohol. Most transgenic research with potato has been directed toward improving yield as well as disease and pest resistance (Doreste et al., 2002; Gulina et al., 1994; Hausler et al., 2002), rather than improving protein quality. Potato is not only protein deficient but also low in lysine, tyrosine, and sulfur amino acids (Jaynes et al., 1986). Consequently, potato is a good candidate for protein improvement by genetic engineering. The possibility of using the BNA to enhance the sulfur content of potato has been investigated (Tu et al., 1998). The CaMV 35S promoter was used to confer constitutive expression of the gene, and this resulted in modest levels of the protein in leaves and tubers. Significantly, it was possible to modify the variable region of the BNA gene so that the protein contains an even higher proportion of methionine. Furthermore, since the allergenicity of the protein appears to reside in this region, it may ultimately be possible to engineer nonallergenic versions of this protein (Tu et al., 1998). The sulfur-rich maize δ-zein has also been expressed in potato tubers, resulting in a substantial increase in sulfur amino acid levels (Li et al., 2001).

The gene encoding the storage albumin from Amaranthus hypochondriacus (AmA1) provides another potential mechanism to increase protein quality (Raina and Datta, 1992). This protein has a good balance of all the essential amino acids and apparently is nonallergenic. AmA1 was expressed in potato under control of the CaMV 35S promoter and the tuber-specific, granule-bound starch synthase (GBSS) promoter, both of which resulted in substantial increases in all essential amino acids in the tubers (Chakraborty et al., 2000). The most highly expressing transgenic lines showed a 2.5- to 4-fold increase in tuber lysine, tyrosine, methionine, and cysteine levels, whereas the GBSS lines had a 4- to 8-fold increase in these amino acids. These changes did not result in the depletion of endogenous proteins (Chakraborty et al., 2000). Consequently, transgenic expression of the AmA1 gene is a promising approach for improvement of protein quality in grain and nongrain crops.

The foliage of pasture crops is also a target for methionine enhancement and may provide a more efficient way to enrich the ruminant diet than with seeds, such as the previously described transgenic lupins. Ruminant livestock, such as cattle and sheep, require methionine in their diet. As previously noted, it is particularly important for sheep that require large amounts of sulfur amino acids for wool production. SSA is a good protein to produce in pasture crops because it is resistant to digestion in the rumen, allowing its amino acids to be absorbed in the small intestine. The subterranean clover (Trifolium subterraneum), which is widely cultivated in Australia, has been transformed with a gene encoding the SSA protein modified with an ER retention signal. Transgenic plants accumulated SSA up to 1.2% of total leaf protein (Khan et al., 1996), but the results of sheep feeding trials have not been reported. Similar constructs were introduced into white clover (Trifolium repens), but much lower levels of the transgenic protein were found to accumulate in the leaves (Christiansen et al., 2000).

The methionine-rich maize zein proteins have also been investigated for their ability to raise foliage methionine levels. When the δ-zein gene was constitutively expressed in white clover, the protein accumulated at up to 1.3% of total protein in all the tissues (Sharma et al., 1998). Birdsfoot trefoil (Lotus corniculatus) and alfalfa (Medicago sativa) are two other foliage crops that have been targeted for methionine improvement by transformation with genes encoding β- and γ-zeins (Bellucci et al., 2002). Earlier work showed that expression of β- and γ-zeins in transgenic tobacco leaves led to the colocalization of these proteins in PBs, underlining the effectiveness of exploiting natural zein interactions in accumulating the proteins in transgenic tissues (Bellucci et al., 2000).

Another approach to improve amino acid deficiencies made use of artificial genes designed to correct specific amino acid deficiencies in target tissues. One strategy employed random ligation of mixtures of small oligonucleotides containing a high proportion of codons for methionine and lysine (Yang et al., 1989). The product was a gene encoding a protein without any clearly defined secondary structure, and it was associated with limited protein accumulation in potato tubers (Yang et al., 1989). In an attempt to produce a synthetic protein with defined secondary structure, Keeler et al. (1997) designed 21-base pair oligonucleotides that encode coiled-coil heptad repeats, forming polypeptides containing up to 31% lysine and 20% methionine. Several different polypeptides were produced that contained up to eight heptad repeats. Under control of the soybean β-conglycinin promoter, this gene resulted in significant increases in lysine and methionine in tobacco seeds that were stable over three generations (Keeler et al., 1997). Such tailor-made proteins are potentially interesting tools for improving the protein quality of seed and nonseed crops, but it remains to be seen whether they would be acceptable to consumers.