Resumen de ELF3 regulates cotyledon expansion and hypocotyl growth rhythmicity
Resumen.
Phytochrome Interacting Factors (PIFs) play a key role in hypocotyl elongation by integrating light, clock and hormonal signals to mediate rhythmic growth (Leivar and Quail 2011). These transcription factors bind G-box (CACGTG) and E-box (CANNTG) elements in the promoters of a diverse set of genes with auxin-related and cell wall-remodeling function and activate their expression (Castillon, Shen et al. 2007). Increased expression of these genes contributes to loosen cell wall, an essential step for elongation growth. Transcriptional activity of these factors is tightly regulated, to prevent excessive growth. Light is the most prominent cue regulating their activity, as it induces rapid destabilization of these factors, via interaction with photoactivated PHYB. This promotes PIFs phosphorylation and marks these factors for proteasomal degradation (Castillon, Shen et al. 2007). GA signaling plays an additional role in the PIFs regulation, with DELLAs being reported to bind the bHLH region of these factors to block DNA recognition ability (de Lucas, Daviere et al. 2008). However, how light and GA signals are coordinated to restrain PIFs activity to the end of the night, hence determining the rhythmic growth pattern observed under SD photoperiodic conditions remains largely unclear.
Interestingly, in a yeast two hybrid screen using the GAI protein as bait, the clock ELF3 protein was uncovered as a new interacting partner of these repressors. This protein, exclusive of plants, plays an important role in the circadian clock by feed-back loop regulating different core components of the clock. Rhythmic seedlings growth depends on clock function, as mutants in the clock core components display an elongated phenotype associated to arrhythmic growth (Nozue, Covington et al. 2007; Niwa, Yamashino et al. 2009). However, the molecular mechanisms underlying this regulation poorly understood at the time this work was initiated. The finding that DELLAs interact with the clock-related ELF3 protein was very rewarding, as pointed to a possible mechanism for rhythmic growth regulation, thus connecting the central oscillator to elongation growth. The elongated phenotype of elf3 mutants actually points to a role of this protein in growth repression, indicating that ELF3 might contribute to DELLA stabilization or to the repression of PIFs activity.
In this work we have found that mutations in the pif4pif5 genes suppress the elongated phenotype of elf3-8 mutants, these factors therefore acting downstream of ELF3. Over-expression of the ELF3 protein, in turn, suppresses the elongated phenotype of PIF4ox lines, indicating that ELF3 functions as a negative regulator of these factors. Expression of the PIF4 was found to be misregulated in elf3-8 seedlings, with increased levels of these transcripts observed at early night. During this interval of the day, growth is usually restricted (Nozue, Covington et al. 2007), elevated expression of these factors hence explaining the tall hypocotyl and arrhythmic growth phenotypes of elf3-8 mutants. Although ELF3 was recently reported to recruit the ELF4 and LUX proteins into the so called evening complex (EC) (Nusinow, Helfer et al. 2011), that represses PIF4/PIF5 expression at dusk, we observed that the PIF4 factor directly interact with the ELF3 protein. This interaction is mediated by the C-ELF3 and bHLH domains of these proteins. Consistent with ELF3 interaction with the PIF4 DNA recognition domain, we established that ELF3 inhibits transcriptional activity of this factor during the day. Although PIFs are widely accepted to be stable only in the dark, we obtained evidence showing that the PIF4 protein starts to accumulate late in the night to peak during the day, although this protein is transcriptionally active only during late night. ELF3 contributes to inhibit PIF4/PIF5 activity during the day, this protein not only repressing PIFs activity as part of the EC complex but also via protein-protein interaction with these factors. Moreover, we observed that PIF4 modulates ELF3 stability, possibly through PHYB-mediated stabilization of this protein. ELF3 is actually reduced in PIF4ox lines and levels of this protein are elevated in the pifQ quadruple mutant, evidencing a complex regulation of these proteins and raising the possibility that PIF4/PIF5 activity feeds-back into the clock by modulating activity of the ELF3 negative loop. Although effects of this regulation were found to be mild in seedlings, in adult plants over-expressing the PIF4 and ELF3 proteins show an intermediate phenotype to ELF3 and PIF4 over-expressers. These seedlings show increased vigor and a prolonged life cycle and are not delayed in floral transition, thus raising a potential agronomic interest of this interaction.
On the other hand, by further analyzing the interaction of ELF3 and DELLAs we have shown that this interaction is mediated by the M-ELF3 domain of ELF3 and the first leucine heptad repeat of DELLAs. Interestingly, M-ELF3 related domains were found to be present also in additional proteins and to be sufficient for DELLA interaction, hence identifying this region as a novel DELLA-interaction domain. As ELF3 had been reported to function as an adapter for COP1 interaction (Yu, Rubio et al. 2008), we analyzed levels of the RGA protein in these two mutant backgrounds. We observed that this repressor is strongly stabilized in these mutants in the dark. However, whereas high levels of the RGA protein can be detected in the elf3-8 mutant, independently of the light-dark cycling conditions, RGA accumulates in cop1-4 seedlings after transition to dark, but activity of this E3 ligase is required for stabilization of this repressor upon transition to light. Noteworthy, high levels of the RGA repressor in elf3-8 mutants contradicted the tall phenotype of these mutants. We show that stabilization of this repressor leads to misexpression of GA-signaling and GA-metabolic genes in this mutant background. Moreover, by generating lines expressing the pRGA::GFP-RGA construct in these genetic backgrounds, we show that RGA accumulates in all plant organs in cop1-4 seedlings, but that in elf3-8 and phyB seedlings this repressor mostly accumulates in the cotyledons. Interestingly, a preferential accumulation of the RGA repressor in the hypocotyl is observed in PHYBox plants, with increased levels of the RGA protein correlating with the hypocotyl lengths and cotyledon sizes of these different plants. Thus, ELF3 and PHYB appear to play an important role in modulating RGA levels in the cotyledons and in this manner its expansion, while COP1 would have a more general effect in all the plant. However, further studies will be required to clarify the exact function of these proteins in modulating RGA stability, keeping in mind that ELF3 interacts with the PHYB, COP1 and DELLA proteins, and that COP1 is required for DELLA stabilization upon transition to light. Altogether, these results evidence a novel mechanism restricting action of the DELLA repressors to specific tissues, unraveling a prevalent role of light in the control of stability of these repressors, independently of GA signaling.
