Metabolism of the Aspartate Family Amino Acids in Developing Seeds: A Balance Between Synthesis and Catabolism

Genetic engineering approaches possess the advantage that gene manipulation can include coding regions as well as regulatory elements such as promoters. Hence, to study the regulation of lysine and threonine metabolism specifically in developing seeds, the E. coli feedback-insensitive AK and DHPS enzymes were expressed in transgenic plants under the control of a seed-specific promoter derived from a gene encoding a seed storage protein. The choice of a storage protein gene promoter was based on the assumption that lysine biosynthesis is spatially and temporally coordinated with storage protein production during seed development. Whether storage protein gene promoters are the best choice to manipulate amino acid metabolism specifically in developing seeds is still unknown and awaits detailed studies of seed development. The first studies included the seed-specific expression of the bacterial feedback-insensitive AK and DHPS in transgenic tobacco plants. Expression of the bacterial AK resulted in significant elevation in free threonine in mature seeds (Karchi et al., 1993), but no increase in free lysine was evident in mature seeds of transgenic plants expressing the bacterialDHPS (Karchi et al., 1994). Developing seeds of these transgenic plants also possessed over tenfold higher activity of lysine-ketoglutarate reductase (LKR), the first enzyme in the pathway of lysine catabolism (Galili et al., 2001), suggesting that the low lysine level in mature seeds of the transgenic tobacco plants resulted from enhanced lysine catabolism (Karchi et al., 1994).

To study the significance of lysine catabolism in regulating free lysine accumulation in seeds under conditions of regulated and deregulated lysine synthesis, Galili and associates have isolated an Arabidopsis T-DNA knockout mutant lacking lysine catabolism (Zhu et al., 2001). This knockout mutant was crossed with transgenic Arabidopsis plants expressing a bacterial feedback-insensitive DHPS in a seed-specific manner (Zhu and Galil, 2003). Although both parental plants contained slightly elevated levels of free lysine compared to wild type in mature seeds, combining both traits into the same plant synergistically boosted free seed lysine levels by ~80-fold, rendering lysine as the most prominent free amino acid (Zhu and Galil, 2003). Moreover, total seed lysine in these plants was nearly doubled compared to wild-type plants (X. Zhu and G. Galili, unpublished results). Notably, the dramatic increase in free lysine in seeds expressing the bacterial DHPS but lacking lysine catabolism was associated with a significant difference in the levels of several other amino acids. The most pronounced differences were significant reductions in the levels of glutamate and aspartate and a dramatic increase in the level of methionine (Zhu and Galil, 2003), exposing novel regulatory networks associated with AAAM and the aspartate family pathway.

A feedback-insensitive DHPS derived from Corynebacterium glutamicum was expressed in a seed-specific manner in two additional transgenic dicotyledonous crop plants, soybean and rapeseed (Falco et al., 1995; Mazur et al., 1999). Seeds of these transgenic plants accumulated up to 100-fold (rapeseed) and several hundred-fold (soybean) higher free lysine than wild-type plants, values that are significantly higher than those obtained in transgenic tobacco plants expressing the E. coli DHPS (Karchi et al., 1994). Whether this is due to the different plant species or to the different bacterial DHPS enzymes is still not clear, but seeds of the lysine-overproducing soybean and rapeseed plants also contained significantly higher levels of lysine catabolic products than wild-type nontransformed plants (Falco et al., 1995; Mazur et al., 1999).

In contrast to dicotyledonous plants in which storage protein synthesis typically takes place in the developing embryo, the synthesis of storage proteins in cereal seeds occurs mostly in the endosperm (Shotwell and Larkins, 1989). Also, based on in situ analysis, the lysine catabolism pathway was suggested to function mostly in the outer layers of the cereal endosperm (Kemper et al., 1999). It is thus expected that expression of a bacterial DHPS, under control of an endospermspecific storage protein gene promoter, will result in enhanced lysine production and perhaps also accumulation of catabolic products of lysine. This expectation was found to be incorrect because lysine overproduction in transgenic maize seeds was observed only when the bacterial DHPS was expressed under an embryo-specific, but not an endosperm-specific promoter (Mazur et al., 1999). Whether the lack of increase in lysine levels upon expressing the bacterial DHPS in the endosperm tissue is due to factors associated with either lysine synthesis or catabolism or both provides an interesting topic for future research.

What then are the functions of lysine catabolism during seed development and why is this pathway stimulated by lysine? The fact that seeds of transgenic soybean, rapeseed, and Arabidopsis can accumulate very high levels of free lysine without a major negative effect on seed germination (only extreme lysine accumulation retards germination) (Falco et al., 1995; Mazur et al., 1999) suggests that lysine catabolism is not required to reduce lysine toxicity. Also, these studies show that the flux of lysine synthesis in developing seeds can become very extensive when the sensitivity of DHPS activity to lysine is eliminated. It is thus possible that during the onset of seed storage protein synthesis, lysine catabolism and likely other amino acids catabolic pathways are stimulated to convert excessfree lysine and other amino acids into sugars and lipids, and also back into glutamate in the case of the lysine catabolism pathway.

The significant research advances in the regulation of lysine metabolism in plants has made this pathway a major biotechnological target for improving the nutritional quality of crop plants. Indeed, a high-lysine corn variety (MaveraTM, Monsanto Inc., St. Louis, Missouri), obtained via embryo-specific expression of a bacterial feedback-insensitive DHPS, has recently been approved for commercial growth for livestock feeding. It is highly likely that additional varieties with higher seed lysine content in which lysine catabolism is reduced and lysine-rich proteins are expressed specifically in seeds will appear in the near future.