Final Report Summary - DEHICIS (Dynamic and evolution of cis-regulatory elements involved in fruit development in Arabidopsis and Brassica)
We studied fruit development both in Arabidopsis and in species with agronomical interest such as oilseed rape. The three major axes of the research we developed are:
1. To understand how networks of transcription factors and their target sequences control flower and fruit development.
We have focused on the regulatory network that controls development of tissues involved in fruit opening, a process that is economically relevant in Brassica because of seed losses caused by premature pod shattering. We found that the gibberellin biosynthesis gene GA4 is a direct target of IND. GA4 was required for proper development of the separation layer and accordingly, ga4 mutants had increased resistance to pod shattering. Similar results were obtained with localised inactivation of gibberellin catalysed by GA2-oxidase expressed from the IND promoter. The defects in fruit development and pod opening seen in ga4 mutants resembled those seen in mutants for the ALCATRAZ (ALC) gene, which encodes a transcription factor that functions downstream of IND. Consistent with a role for gibberellin in promoting ALC function, the ALC protein interacted directly with DELLA transcriptional repressors, which are destabilised by gibberellin. In addition, the defects caused by low gibberellin were reversed by loss of DELLA function. The results were published in a paper and served as the basis for a patent application.
2. To understand how gene networks (relevant to flower and fruit development) vary between species.
We aimed to identify cis-regulatory elements in RPL, and investigate whether variation in these regulatory sequences underpins differences in fruit development and pod shattering seen between Arabidopsis and Brassica. Using phylogenetic footprinting, we found that mutation of a single conserved cis-element in RPL is correlated with variation in replum morphology. Reciprocal transformation experiments confirmed that this single nucleotide change was the cause of the differences in replum morphology between Arabidopsis and Brassica. A mutation in the same nucleotide position of the same cis-element in RPL has been independently selected during rice domestication to reduce seed shattering since rice varieties carrying this mutation have defects in grain abscission. Together these results provide an exemple of repeated evolution underlied by genetic convergence and suggest that the same genetic toolkit is relevant to domestication and natural evolution in widely diverged species.
3. To explore variations in the flower/fruit gene regulatory network for practical use.
One of the aims of this project was to use our knowledge of cis-regulatory elements to isolate regulatory mutants with changes in fruit development leading to reduced pod shattering in Brassica. JAG is a good candidate gene: it has been shown that mis-expression of JAG in the Brassica fruit abolishes replum formation and prevents pod opening. We have identified a conserved cis-element that restricts expression of the JAGGED gene and used TILLING to isolate a Brassica JAG2 allele with this element mutated. The mutant plants had increased JAG2 expression and showed phenotypes previously described for gain of JAG function (which in Arabidopsis includes reduced seed loss). This is a proof of principle that targeted isolation of cis-regulatory mutants in Brassica is feasible, although further work is required to test whether the new JAG2 allele will be useful in breeding for reduced pod shattering.
1. To understand how networks of transcription factors and their target sequences control flower and fruit development.
We have focused on the regulatory network that controls development of tissues involved in fruit opening, a process that is economically relevant in Brassica because of seed losses caused by premature pod shattering. We found that the gibberellin biosynthesis gene GA4 is a direct target of IND. GA4 was required for proper development of the separation layer and accordingly, ga4 mutants had increased resistance to pod shattering. Similar results were obtained with localised inactivation of gibberellin catalysed by GA2-oxidase expressed from the IND promoter. The defects in fruit development and pod opening seen in ga4 mutants resembled those seen in mutants for the ALCATRAZ (ALC) gene, which encodes a transcription factor that functions downstream of IND. Consistent with a role for gibberellin in promoting ALC function, the ALC protein interacted directly with DELLA transcriptional repressors, which are destabilised by gibberellin. In addition, the defects caused by low gibberellin were reversed by loss of DELLA function. The results were published in a paper and served as the basis for a patent application.
2. To understand how gene networks (relevant to flower and fruit development) vary between species.
We aimed to identify cis-regulatory elements in RPL, and investigate whether variation in these regulatory sequences underpins differences in fruit development and pod shattering seen between Arabidopsis and Brassica. Using phylogenetic footprinting, we found that mutation of a single conserved cis-element in RPL is correlated with variation in replum morphology. Reciprocal transformation experiments confirmed that this single nucleotide change was the cause of the differences in replum morphology between Arabidopsis and Brassica. A mutation in the same nucleotide position of the same cis-element in RPL has been independently selected during rice domestication to reduce seed shattering since rice varieties carrying this mutation have defects in grain abscission. Together these results provide an exemple of repeated evolution underlied by genetic convergence and suggest that the same genetic toolkit is relevant to domestication and natural evolution in widely diverged species.
3. To explore variations in the flower/fruit gene regulatory network for practical use.
One of the aims of this project was to use our knowledge of cis-regulatory elements to isolate regulatory mutants with changes in fruit development leading to reduced pod shattering in Brassica. JAG is a good candidate gene: it has been shown that mis-expression of JAG in the Brassica fruit abolishes replum formation and prevents pod opening. We have identified a conserved cis-element that restricts expression of the JAGGED gene and used TILLING to isolate a Brassica JAG2 allele with this element mutated. The mutant plants had increased JAG2 expression and showed phenotypes previously described for gain of JAG function (which in Arabidopsis includes reduced seed loss). This is a proof of principle that targeted isolation of cis-regulatory mutants in Brassica is feasible, although further work is required to test whether the new JAG2 allele will be useful in breeding for reduced pod shattering.