Activator and Repressor E2Fs Who is

As mentioned above, RBR1 function is essential in Arabidopsis, suggesting that plant E2Fs are required for the regulation of cell division and differentiation. However, overexpression of a dominant negative (DN) mutant form of DPA, which affected DNA binding but maintaining the dimerization ability, challenged this notion (Ramirez-Parra et al. 2003; Magyar and De Veylder, unpublished data) (Fig. 2). In these transgenic cells where E2F transcriptional activity is blocked, DELs should be the only remaining E2Fs due to their DP-independent binding to E2F-targeted promoters. Although a similar mutation in mammalian DP-1 induces G1 cell cycle arrest (Wu et al. 1996), in these DPA-DN Arabidopsis plants there were no macroscopic phenotypes or cell cycle defects, just a minor reduction in the expression levels of certain E2F target genes (e.g. RNRII, ORC1) (Ramirez-Parra et al. 2003; Magyar and DeVeylder, unpublished data). These data further support the idea that, in plants, DELs are unlikely to be the major repressors of cell division that antagonize E2F functions. It is possible, however, that the mutant DPA eliminates both the repressor and the activator DP-dependent E2F complexes from the DNA, which resulted in normal development of the transgenic Arabidopsis plants. Mutations of the activator and repressor Drosophila dE2Fs resulted in only a minor defect in cell proliferation and growth during larval development (Frolov et al. 2001). Interestingly, the effects of inactivating the single Drosophila dDP appeared indistinguishable from the effects of inactivating both dE2F1 and dE2F2 (Frolov et al. 2001, 2003). Therefore, the individual Arabidopsis E2Fs might work antagonistically on cell division and growth in a similar way to that demonstrated in Drosophila. Since cell proliferation did not change significantly in the DP mutant plants, it is likely that plant E2F functions are not vital for the transcriptional activation of cell cycle genes, in agreement to what was found with E2F and DP mutants in Drosophila and in Chlamydomonas (Fang et al. 2006; Frolov et al. 2001). However, it cannot be excluded that the mutant DPA protein is not able to eliminate all the DP-dependent E2F functions in the cells, and thus some remaining E2F activities could still drive cells through the cell cycle. Further studies of individual E2F mutants and the analysis of their combinations are required to understand the functions and interactions of E2Fs during plant development, and to fully address whether they are indispensable for regulation of cell cycle transitions.

Functional characterization of the individual members of Arabidopsis E2Fs have already revealed differences among them: ectopic expression of E2FA with DPA resulted in strong activation of both mitotic cell cycle and endo-cycle (De Veylder et al. 2002; Kosugi and Ohashi 2003); overexpression of E2FB was also able to activate mitosis but it repressed the endocycle (Magyar et al. 2005; Sozzani et al. 2006), whereas reduction in the level of E2FC confirmed its negative regulatory function in mitosis but a positive one in endoreduplication (del Pozo et al. 2006). According to these data, E2FB and E2FC are antagonistic transcription factors, while E2FA has a dual functionality. We do not know much about the target gene specificities of E2FB and E2FC, but there is a clear antagonistic effect of E2FB and E2FC on the regulation of the mitotic cyclin B1;1 gene (del Pozo et al. 2006; Sozzani et al. 2006). Although the promoter of this cyclin gene does not contain consensus E2F binding elements, decreasing the amount of E2FC in transgenic Arabidop-sis plants, by using E2FC-specific RNAi, dramatically activated its expression (90 times up), even to a much higher level than the expression of known E2F target genes such as CDC6, EXP3. This indicates that this mitotic cyclin could be a direct target for E2FC-dependent repression. Since overexpression of E2FB, or co-expression of E2FA with DPA, significantly activates the expression of cyclin B1;1, it is also possible that it has an alternative E2F-binding site; however, this needs to be tested experimentally. In contrast, these data strongly support a G2-M regulatory function for plant E2Fs, similarly to animal E2Fs (Hernando et al. 2004; Neufeld and Edgar 1998). Furthermore, increased E2FB levels led to shortened cell cycle duration, and it was suggested that E2FB function in plant cells is comparable to that of Drosophila dE2F1, which simultaneously increases the expression of critical S- and M-phase regulators (Magyar et al. 2005).

Our recent data show that E2FB can directly induce the promoter of the Arabidopsis CDKB1;1 gene, a plant-specific regulator of the G2 to M transition (Magyar and Bogre, unpublished results). It is important to note that reduction in the level of E2FC caused significant changes in gene expression in mature leaves but not in young leaves, suggesting that E2FC acts by repressing E2F-regulated genes in mature differentiated cells. As the ploidy level decreased in these transgenic plants, it can be argued that E2FC has an important regulatory role in the switch from mitotic cell cycle to endocycle by repressing the expression of mitotic genes such as cyclin B1;1. Interestingly, in switching from mitotic to endocycles, cells in Dorosophila embryo terminate the expression of mRNAs encoding the mitotic regulators such as cyclin B1 and B3 genes (Edgar and Orr-Weaver 2001). Ectopic expression of mitotic cyclin B1;2 in Arabidopsis trichome cells switch the endocycle into mitosis, indicating that reducing the expression of mitotic cyclins is an important regulatory step towards the activation of endocycle (Schnittger et al. 2002). How E2FC regulates the transcription of cyclin B1;1 is not known yet, but previous studies indicated that E2FC could work as a direct transcriptional repressor on E2F target genes.

Ectopic expression of a stabilized mutant AE2FC lacking the amino terminal domain in dark-grown Arabidopsis plants resulted in reduced expression of CDC6, an S-phase regulatory gene (del Pozo et al. 2002) (Fig. 2). Intrigu-ingly, the two alternative splicing variants of the mouse E2F3, the natural full length (E2F3a) and the amino terminally deleted (E2F3b) forms displayed opposite transcriptional functions where the activator role of E2F3a is converted into transcriptional repressor in E2F3b (Aslanian et al. 2004; Leone et al. 2000). Possibly the analogous deletion on Arabidopsis E2FC might have similar functional impacts (Fig. 2). Although the precise role of the N-terminal region of mammalian E2F1-3 is still being uncovered, it is required for ubiquitin-mediated degradation of E2F1 protein (Marti et al. 1999). The N-terminal extension of plant E2Fs contains a number of CDK phospho-rylation sites (Fig. 2) and the E2FC protein was targeted by the ubiquitin-mediated proteosome pathway in a CDK-dependent manner. Furthermore, deletion of the N-terminal part stabilized the E2FC as well as E2FA protein (del Pozo et al. 2002; Magyar 2005; Magyar and Bogre, unpublished). Although clear evidence that E2FC works as a direct transcriptional repressor on E2F target genes is still missing. The strong growth-arrested phenotype of the N-terminal deletion AE2FC mutant co-expressed with DPB dimeriza-tion partner was interpreted as a consequence of reduced cell proliferation due to abundance of E2FC-DPB transcriptional repressor complex (del Pozo et al. 2006). The higher ploidy level observed in the double transgenic mutant AE2FC-DPB further supported the hypothesis that E2FC is a negative regulator for mitosis but an activator of endocycle.

It would be interesting to know whether E2FC has a role in meristems or in young leaves, since its mitosis inhibitory function seems to be restricted to mature leaves, while E2FC expression was found to be high in actively dividing tissues (del Pozo et al. 2002, 2006). Microarray studies indicate constitutive expression in both leaves and roots (Beemster et al. 2005). Ectopic expression of E2FA with DPA also resulted in a strong growth-arrested phenotype in transgenic Arabidopsis plants (De Veylder et al. 2002) or inhibited tobacco growth in a concentration-dependent manner (Kosugi and Ohashi 2003). Cell proliferation and endoreduplication, however, were both strongly up-regulated in E2FA/DPA overexpressors. How E2FA could control in parallel these different, spatially and temporally separated processes has not yet been addressed. E2FA transcripts were detected both from mitotic and en-doreduplicating tissues, indicating that E2FA could have a dual regulatory role in vivo in both of these events. Moreover, it was found that ectopic E2FA has an opposite effect on cell proliferation during Arabidopsis development. It increased cell number in the cotyledons (De Veylder et al. 2002) but resulted in fewer cells in mature leaves (He et al. 2004). Probably, E2FA could activate or repress cell division depending on the developmental stage. In addition, increased E2FA activity in E2FA/DPA co-expressors resulted in almost complete inhibition of growth early after germination, probably due to arrest of cell cycle exit (De Veylder et al. 2002).

Genome-wide expression analysis of these transgenic Arabidopsis plants revealed that genes encoding proteins required for DNA synthesis are highly up-regulated (e.g. CDC6, ORC1, CDC45, RNRII, MCM3), strongly supporting a regulatory role for E2FA during DNA synthesis. Moreover, G2-M regulatory genes such as cyclin B1;1, and CDKB1;1 were also found among the potential

E2FA target genes (Vandepoele et al. 2005; Vlieghe et al. 2003). Promoter analysis of CDKB1;1 further supported the notion that the E2FA-DPA heterodimer could activate this G2-M specific promoter through its E2F binding element (Boudolf et al. 2004). However, these results raise the questions on how E2FA could stimulate G2 to M transition when its own expression is restricted to the S-phase of the cell cycle, and how E2FA could activate endocycle if it is a positive transcriptional regulator ofgenes required for G2 to M transition.

Intriguingly, E2FA seems to negatively regulate cell proliferation in mature leaves, as observed for E2FC, but the molecular mechanism behind this ob-

CYCD/CDKA

G1 to S phase

S phase G2 to M phase

Endocycle/Differentiation

Fig. 3 Model of Arabidopsis RBR1-E2F pathway. The model reflects our current understanding of how RBR1 and E2Fs control cell proliferation and differentiation. Growth-promoting factors such as light and auxin activate RBR-kinase complexes, consisting of cyclin D or cyclin A and CDKA;1 that inactivate RBR1 function by hyperphosphorylation. CDK inhibitors (KRPs) suppress RBR1 phosphorylation resulting in hypophosphorylated RBR1. The single Arabidopsis RBR1 protein is able to interact with all the three E2Fs and thus repress their transactivation functions, although the mechanism of this interaction is still unknown. E2FB and E2FC have opposing functions: E2FB activates both G1-S and G2-M cell cycle transitions, while it inhibits endoreduplication. In contrast, E2FC stimulates the switch from mitosis to endocycle by repressing G2-M transition. Growth-promoting factors such as light and auxin stabilize E2FB, but destabilize E2FC proteins. E2FA is a strong activator for the S-phase of the cell cycle, but it also could stimulate mitosis and endocycle through the up-regulation of E2FB and E2FC expression. E2FC and E2FB could negatively regulate E2FA expression and protein accumulation, respectively, by unknown feed-back mechanisms (del Pozo et al. 2006; Sozzani et al. 2006)

G1 to S phase

S phase G2 to M phase

Endocycle/Differentiation

Fig. 3 Model of Arabidopsis RBR1-E2F pathway. The model reflects our current understanding of how RBR1 and E2Fs control cell proliferation and differentiation. Growth-promoting factors such as light and auxin activate RBR-kinase complexes, consisting of cyclin D or cyclin A and CDKA;1 that inactivate RBR1 function by hyperphosphorylation. CDK inhibitors (KRPs) suppress RBR1 phosphorylation resulting in hypophosphorylated RBR1. The single Arabidopsis RBR1 protein is able to interact with all the three E2Fs and thus repress their transactivation functions, although the mechanism of this interaction is still unknown. E2FB and E2FC have opposing functions: E2FB activates both G1-S and G2-M cell cycle transitions, while it inhibits endoreduplication. In contrast, E2FC stimulates the switch from mitosis to endocycle by repressing G2-M transition. Growth-promoting factors such as light and auxin stabilize E2FB, but destabilize E2FC proteins. E2FA is a strong activator for the S-phase of the cell cycle, but it also could stimulate mitosis and endocycle through the up-regulation of E2FB and E2FC expression. E2FC and E2FB could negatively regulate E2FA expression and protein accumulation, respectively, by unknown feed-back mechanisms (del Pozo et al. 2006; Sozzani et al. 2006)

servation is not known (He et al. 2004). It has been reported that E2FA is able to activate the expression of both E2FB and E2FC, the two potentially antagonistic E2F transcription factors on the regulation of mitosis and endocycle, respectively (Vandepoele et al. 2005). Therefore, it is possible that the balance between E2FB and E2FC levels will determine whether the cells continue in mitosis or switch to endocycle. E2FA has an effect on both of these factors through the regulation of E2FB and E2FC. How the ratio between E2FC and E2FB is set is not known, but an interesting hypothesis suggests that auxin distribution plays a role in this process.

Auxin regulates cell division and elongation in a concentration-dependent manner; elevated auxin levels activate cell division in the meristems, while reduced amounts repress mitosis as cells leave the meristematic regions, and in parallel it enhances cell growth (Scheres and Xu 2006). Auxin increases the stability of E2FB, and co-expression of E2FB with DPA in plant cells could maintain cell proliferation even in the absence of auxin (Magyar et al. 2005). Moreover, elevated levels of E2FB-DPA heterodimer in plant cells resulted in extremely small cell size, indicating that E2FB inhibits growth. Ectopic expression of E2FB in Arabidopsis plants also stimulates cell division and results in smaller cells both in leaf and in roots that became significantly shorter (Sozzani et al. 2006). It was suggested that auxin could influence cell proliferation and growth through the modulation of the level of E2FB protein; a high auxin level would stabilize E2FB, which stimulates cell division. In contrast, E2FC stability was shown to be oppositely regulated, destabilized in growth-promoting physiological conditions (e.g. in plants grown in light), and regulated by the ubiquitin-SCF pathway (del Pozo et al. 2002). Therefore, an elevated level of E2FB could specifically keep cells dividing in the meristems and in young tissues, while the E2FC protein level would increase above E2FB in mature leaves and thus would stimulate the switch from mitotic cell cycle to endocycle by repressing mitotic genes (Fig. 3).

Was this article helpful?

0 0

Post a comment