Metabolic Fluxes of the Aspartate Family Pathway Are Regulated by Developmental, Physiological, and Environmental Signals

Although the aspartate family pathway is subject to major regulation by feedback inhibition loops, the fluxes of this pathway also depend on the expression of genes encoding the enzymes of this pathway. Expression of the genes and activities of the encoded enzymes may be regulated by transcriptional, posttranscriptional, translational, and posttranslational mechanisms, which may respond to various developmental, physiological, and metabolic signals. One way to identify such regulatory signals is to test their effects on the steady-state levels of the aspartate family amino acids and on the expression and activity of enzymes of this pathway. However, since the aspartate family amino acids are relatively minor amino acids (Noctor et al., 2002), it is difficult to draw statistically meaningful conclusions from such studies. Hence, metabolic engineering of feedback inhibition loops appears to be the appropriate strategy for functional dissection of signals that regulate the production of the aspartate family enzymes as rationalized in the following. Although feedback inhibition of DHPS and AK represents major regulators of the fluxes of the aspartate family pathway, synthesis of its end-product amino acids also depends on the expression of additional enzymes in this pathway (Fig. 3.2). Thus, if a feedback-insensitive DHPS or AK were to be constitutively expressed in transgenic plants, significant lysine or threonine overproduction would be expected only in the specific tissues or growth conditions where the genes encoding the entire set of lysine and/or threonine biosynthetic enzymes are also abundantly expressed. Indeed, lysine levels in transgenic plants constitutively expressing a feedback-insensitive bacterial DHPS fluctuated considerably under different growth conditions, being higher in young leaves and floral organs than in old leaves, and positively responding to light intensity (Shaul and Galili, 1992a; Zhu-Shimoni and Galili, 1998). In contrast, threonine levels in transgenic plants constitutively expressing a bacterial feedback-insensitive AK showed much less fluctuations than lysine levels in plants expressing the E. coli feedback-insensitive DHPS (O. Shaul and G. Galili, unpublished information). The results imply that metabolic fluxes of the aspartate family pathway are regulated by developmental, physiological, and environmental signals and that fluxes in the lysine and threonine branches respond differently to the signals.

The regulation of synthesis of the aspartate family amino acids was studied further by analyzing the expression patterns of two Arabidopsis genes encoding AK/HSD and DHPS enzymes, using Northern blot analyses and promoter fusion to the β-glucuronidase (GUS) reporter gene. The developmental expression pattern of both genes was very similar, that is, they were highly expressed in germinating seedlings, actively dividing and growing young shoot and root tissues, various organs of the developing flowers, as well as in developing embryos (Vauterin et al., 1999; Zhu-Shimoni et al., 1997). Exposure of etiolated seedlings to light results in an altered pattern of GUS staining in the hypocotyls and cotyledons, suggesting that expression of the AK/HSD and DHPS genes is also regulated by light (Vauterin et al., 1999; Zhu-Shimoni et al., 1997). This was supported by studies showing that the levels and activities of the barley AK isozymes are increased by light and phytochrome (Rao et al., 1999). The similarities in the developmental and light-regulated patterns of expression of the AK and DHPS genes suggest some coordination of expression of genes encoding enzymes of the aspartate family pathway. However, this clearly does not account for the entire set of the aspartate family genes as deduced from the differential expression pattern of two of the three Arabidopsis genes encoding lysine-sensitive monofunctional AK isozymes. Based on an analysis of transgenic plants expressing promoter-GUS constructs, expression of one of these genes was more predominant than the other in vegetative tissues (Jacobs et al., 2001). Both genes were highly expressed at the reproductive stage, but only one of these genes was expressed in fruits (Jacobs et al., 2001). Whether this variation in expression pattern reflects a nonredundant function of the different AK isozymes or association with developmentally regulated variations in metabolic fluxes of the lysine and threonine branches, discussed above, remains to be elucidated.