SeeDev-binary@ldeleger:SeeDev-binary-19719479-3 JSONTXT

WRI1 directly enhances the expression of genes involved in glycolysis and fatty acid biosynthesis Factors controlling the overall level of oil stored in seeds of A. thaliana, and the integration of this biochemical process in the complex framework of seed development, are only partially characterized (Santos Mendoza et al., 2008). Complementary expression analyses and transcriptomic approaches have demonstrated that the biosynthetic pathways for fatty acids and lipids are largely regulated at the transcriptional level in maturing seeds (Ruuska et al., 2002). The ‘contrapuntal’ or differential timing of the expression of genes involved in oil metabolism during seed development most probably reflects distinct regulatory pathways controlling the transcription of different sets of genes involved in fatty acid synthesis and TAG assembly, respectively. As groups of enzymes within a pathway sometimes exhibit similar regulation, it has been postulated that those members of a given pathway might share common cis- and trans-regulatory elements. For instance, a number of genes encoding enzymes of the glycolysis and the fatty acid biosynthetic pathway display a bell-shaped pattern of expression between 5 and 13 DAA (Ruuska et al., 2002). This is corroborated in our study by qRT-PCR experiments aimed at characterizing the steady-state accumulation of BCCP2 and PKp-β1 mRNAs among a developmental series of maturing siliques (Figure 3a) or seeds (Figure S1). The WRI1 transcription factor, previously shown to represent a node in the regulatory network controlling oilseed metabolism (Cernac and Benning, 2004), was proposed to trigger the transcription of the set of genes involved in the conversion of sucrose into fatty acids. This hypothesis was based on the analysis of the expression profiles of putative target genes of WRI1 characterized in various wri1 mutant backgrounds, and in tissues overexpressing WRI1 ectopically (Ruuska et al., 2002; Baud et al., 2007a). In this study, the use of the uidA reporter gene, the expression of which is driven by full-length promoter sequences of either BCCP2 or PKp-β1, further establishes that WRI1 is able to strongly modulate the activity of these two promoters in planta (Figure 4a,b). Finally, EMSA (Figure 8) and yeast one-hybrid experiments (Figure 7a) demonstrate that WRI1 is able to interact with the BCCP2 promoter, strongly suggesting that this gene is a direct target of WRI1 in planta. It is noteworthy that the fine characterization of the changes in mRNA accumulation for BCCP2, PKp-β1 and WRI1 throughout silique development shows that the induction of the regulatory factor precedes the induction of its targets in the Col-0 accession (Figure 3). This slight delay in the detection of mRNAs corresponding to target genes of WRI1 may presumably illustrate biochemical processes occurring in embryo cells, such as translation of the WRI1 peptide, and post-translational regulations (e.g. translocation of WRI1 towards the nucleus or putative recruitment of its interacting partners). Whereas Cernac and Benning (2004) have reported that expression of WRI1 cDNA under the control of the cauliflower mosaic virus 35S-promoter leads to slightly increased seed oil content, neither the Pro35Sdual:WRI1 nor the ProS2:WRI1 transgene used in the present study efficiently stimulates oil accumulation in the corresponding transgenic seeds (Tables 1 and 2). Several factors like growth conditions or strength and specificity of the different promoters used (classical Pro35S versus Pro35Sdual or ProS2) may explain these apparent discrepancies. However, it is also interesting to note that the transgenes considered have been introduced in different A. thaliana accessions (Col-0 versus Ws backgrounds). In the Ws background (this study), the endogenous expression level of glycolytic and fatty acid biosynthetic genes is significantly higher than in the Col-0 accession (Figure S1). Likewise, fatty acid concentration is more elevated in mature dry seeds of the Ws accession (see Figure 1c in Baud et al., 2007a). Therefore, it is tempting to speculate that expression of lipogenic genes may be more limiting in a Col-0 background than in a Ws background.