PerspectivePlant biology

Advances in Understanding Brassinosteroid Signaling

See allHide authors and affiliations

Science's STKE  26 Sep 2006:
Vol. 2006, Issue 354, pp. pe36
DOI: 10.1126/stke.3542006pe36

Abstract

Brassinosteroids (BRs) function as signaling molecules in plants and are involved in processes such as stem elongation, vascular differentiation, male fertility, timing of senescence and flowering, leaf development, and resistance to biotic and abiotic stresses. Unlike animal steroids that are perceived by nuclear receptors, BRs are perceived by transmembrane receptor kinase complexes that initiate a phosphorylation-mediated signaling cascade to transduce the steroid signal. BR binding to the extracellular domain of the receptor BRI1 induces kinase activation and hetero-oligomerization with the second transmembrane kinase BAK1. Activated BRI1 then dissociates from the BRI1-interacting protein BKI1, a newly identified negative regulator of BR signaling. In the presence of BR, the kinase BIN2, which is the Arabidopsis homolog of GSK3 (glycogen synthase kinase 3), is inhibited by an unknown mechanism, leading to dephosphorylation of BES1 and BZR1 inside the nucleus. This allows BES1 and BZR1 to homodimerize or combine with other transcription factors to bind to promoters of BR-responsive genes. These studies of BR signaling in plants have revealed signaling pathways that are distinctly different from related ones operating in animal cells.

Steroid hormones play key roles in growth and development of multicellular eukaryotes. Plant steroid hormones are called brassinosteroids (BRs) and regulate cellular expansion, differentiation, and proliferation. BRs are involved in diverse processes, such as stem elongation, vascular differentiation, male fertility, timing of senescence and flowering, leaf development, and resistance to biotic and abiotic stresses (13). Unlike animal steroids that are perceived by nuclear receptors, plant steroids are perceived by transmembrane receptor kinases that initiate a phosphorylation-mediated signaling cascade to transduce the steroid signal. BRs bind to a subdomain within the extracellular leucine-rich repeats (LRRs) of the plasma membrane receptor kinase BRI1 (BRASSINOSTEROID-INSENSITIVE 1), as well as to two of its close relatives BRL1 (BRI1-Like 1) and BRL3 (46). Both genetic and biochemical approaches have begun to provide further details of the various components of the BR signal transduction pathway.

On the basis of genetic screens and yeast-based protein-protein interaction screens, a second LRR receptor–like kinase (RLK) designated BAK1 (BRI1 ASSOCIATED KINASE 1) was identified (7, 8) that most likely does not bind BR itself (5). Two elegant studies of the in vivo phosphorylation properties of BRI1 revealed that after binding of BR to the BRI1 receptor, homodimerization was required (9) to promote hetero-oligomerization with BAK1 (10), which may result in a tetrameric receptor complex (Fig. 1). Auto- or transphosphorylation of the C terminus of BRI1 is instrumental in enhancing the kinase activity of BRI1, as well as promoting its affinity for the coreceptor BAK1 (9, 10). BAK1 is identical to the third member of the SERK (somatic embryogenesis receptor-like kinase) family. After hetero-oligomerization, BAK1 (also known as SERK3) accelerated internalization by endocytosis of BRI1 (11). Another member of the SERK family, SERK1, is also able to interact with BRI1 and appears to modulate BRI1 signaling in a manner similar to that observed for BAK1 (12). Although the precise stoichiometry is not yet known, the emerging picture suggests that SERK and BRI receptors exist as homodimers in the absence of ligands and form ligand-induced tetrameric complexes, analogous to the receptor complexes of mammalian transforming growth factor–β (TGF-β) receptors. Precise combinatorial interactions in the tetrameric TGF-β receptor complex allow differential ligand binding or differential signaling in response to the same ligand (13). Whether this also applies to the BRI1-SERK complexes is not known. BRI1 also interacts with TRIP-1, a cytosolic protein that is homologous to a TGF-β receptor interacting partner (14).

Fig. 1.

Model of the BR signal transduction pathway. The plasma membrane–localized receptor kinase BRI1 is the major BR receptor. (A) Without BR, BRI1 and its coreceptors BAK1 or SERK1, or both, recycle through endosomes. The kinase activity of BRI1 is kept in a basal state by both its own C-terminal domain and by an interaction with BKI1. (B) BR binding to the extracellular domain of BRI1 induces a conformational change of the kinase domain, leading to the phosphorylation of BRI1 and dissociation of BKI1 from the plasma membrane. Following the activation of BRI1, together with BAK1 or SERK1, a multimeric complex is formed. This complex may consist of different members of both BRI and SERK families; only one possible configuration (12) is shown. This activation leads to dephosphorylation of the nuclear proteins BES1 and BZR1 by inhibiting the negative regulator BIN2. BES1 and BZR1 then homodimerize and bind to DNA to regulate BR-responsive genes. In the absence of BR, the kinase BIN2 phosphorylates BES1 and BZR1. BSU1 is a nuclear phosphatase involved in the dephosphorylation of BES1. The entire hetero-oligomeric receptor complex can be internalized by endocytosis. It is not certain whether internalization of the receptor complex requires BR or whether the receptor complex is capable of propagating the BR signal after internalization. The model is a modification of the one presented by Wang and Chory (23).

On the basis of genetic screens, other downstream components of BR signaling have been identified. BIN2 (BR INSENSITIVE 2) is a glycogen synthase kinase 3 (GSK3) homolog. Genetic and biochemical evidence suggested that BIN2 is a negative regulator of BR signaling and functions by phosphorylating and thereby inactivating two nuclear proteins, BZR1 (BRASSINAZOLE RESISTANT 1) and BES1 (BRI1 EMS SUPPRESSOR 1). Both are required for BR-regulated gene expression in a mechanism proposed to be analogous to that of Wnt signaling in fruit fly and vertebrates (1518). When overexpressed in Arabidopsis plants as fusion proteins with the green fluorescent protein (GFP), BES1 and BZR1 are rapidly dephosphorylated after treatment with BR (18, 19), probably by the phosphatase BSU1 (BRI1 SUPPRESSOR PROTEIN 1), a nuclear protein (20) (Fig. 1).

Vert and Chory (21) now extend these observations by showing that BR signaling differs from the previously proposed canonical Wnt signaling mechanism in which the GSK3β protein is inactivated in the presence of the ligand, leading to accumulation and nuclear translocation of the now dephosphorylated β-catenin (22). Individual loss-of-function alleles of BIN2 and two of its closest homologs revealed no effect on BR signaling, but the triple knockout displayed constitutive BR responses and strong resistance to the BR biosynthetic inhibitor brassinazole, suggesting that GSK3 kinases act redundantly to negatively regulate BR signaling. Furthermore, BIN2-GFP was localized in the nucleus, in the cytosol, and at the plasma membrane. In contrast, a GFP fusion with bin2 containing the E263K mutation that produced the original gain-of-function phenotype was found mostly localized to the nucleus. After expressing the mutant bin2 protein fused to a nuclear localization signal, severe dwarf phenotypes resembling the genetic defects observed in the absence of synthesis or perception of BRs were observed. These results showed that BIN2 activity is required inside the nucleus and not in the cytoplasm. The conversion of BES1 proteins from the dephosphorylated to the phosphorylated state was consistent with the observed dwarf phenotypes. After expression in transgenic plants the BES1-GFP protein was constitutively nuclear, which further supports the model that BIN2-mediated phosphorylation of BES1 is exclusively nuclear. Neither protein stabilization nor nuclear translocation of BES1 is apparently required for BR signaling, emphasizing the importance of phosphorylation in regulating BES1 activity. Finally, BIN2-mediated phosphorylation of BES1 resulted in a complete loss of its DNA-binding activity to downstream target promoters, and its ability to multimerize was also clearly decreased. These results indicate that rather then regulating the nuclear translocation and accumulation of BES1, BIN2 acts in the nucleus to regulate the genomic response to BRs through phosphorylation of BES1, which blocks the transcriptional activity of BES1.

In a second study, Wang and Chory describe another negative regulator of BR signaling, BKI1 (BRI1 KINASE INHIBITOR 1) (23). BKI1 interacted directly with the kinase domain of BRI1, as did transthyretin-like (TTL) protein (24), both of which were identified by yeast two-hybrid screening. The C-terminal domain of BKI1 was both necessary and sufficient to bind the BRI1 kinase domain and the interaction was highly specific, because BKI1 did not interact with TTL, BIN2, or the kinase domain of other RLKs. Immunoprecipitation experiments confirmed that BRI1 interacted with a BKI1-FLAG fusion in vivo, and with the use of promoter–β-glucuronidase (GUS) fusions, BKI1 was shown to be coexpressed with BRI1 in a number of plant tissues (23, 25). RNA interference (RNAi) lines of BKI1 showed that BKI1 represses BR-related growth, whereas overexpression of BKI1 resulted in dwarf plants resembling plants harboring weak alleles of bri1. As could be predicted from the phenotype, dephosphorylated BES1 in the BKI1 overexpression lines was almost undetectable, suggesting that BRI1 signaling was suppressed by BKI1 overexpression. The BKI1-YFP (yellow fluorescent protein) fusion was localized both at the plasma membrane and the cytosol. Within 5 min of BR application, the plasma localization of BKI1 was shifted completely to the cytosol. No BR-induced dissociation of BKI1-YFP from the plasma membrane was observed in a bri1 kinase–inactive mutant background (25), suggesting that the kinase activity of BRI1 is required for BKI1-YFP relocalization. On the basis of these findings, a new model is proposed where in the absence of BRs, BRI1 homodimers are kept localized at the plasma membrane in association with BKI1 proteins. This may in fact keep BRI1 from association with its coreceptors, such as BAK1 or SERK1 (Fig. 1).

These new findings are summarized in a model based on various previous ones and shows that BR binding to the extracellular domain of BRI1 induces receptor phosphorylation and activation, as well as its dissociation from BKI1. In the absence of BRs, BIN2 negatively regulates BR signaling by phosphorylating the transcriptional regulators BES1 and BZR1. In the presence of BRs, BIN2 is inhibited by an unknown mechanism, leading to dephosphorylation of BES1 and BZR1, which then homodimerize or cooperate with other transcription factors, which allows DNA binding and regulation of BR-responsive genes (21, 26).

This new signaling paradigm raises several interesting questions. The first is the stoichiometry and conformation of the BR receptor complex in the plasma membrane in the absence and presence of BRs. To what extent are the proposed tetrameric complexes preformed, or are these tetrameric complexes only assembled upon ligand application? Both configurations appear to coexist in heterologous plant cells (11). A second question is whether the complex is the same in all cells where the receptors are expressed or whether there are a multitude of forms composed of different family members. This is important for determining whether the proteins so far identified are capable of integrating or separating the different cellular and developmental effects of BRs.

A third question concerns the role of the coreceptors in BR signaling. One hypothesis, which is modeled on TGF-β signaling, suggests that BR binding activates BRI1, which can then phosphorylate BAK1 to propagate BR signaling (7, 12). This hypothesis is supported by Wang et al. (10), who showed that BR treatment not only stimulates the phosphorylation of both BRI1 and BAK1, but also enhances their physical interaction in transgenic plants coexpressing tagged BRI1 and BAK1. Thus, downstream components intermediate between BRI1 and BIN2 may interact with BAK1 rather then BRI1 itself. However, Karlova et al. (12), using the SERK1 protein as bait to identify interacting proteins, did not identify any known components of BR signaling other than the BRI1 and BAK1 receptors. Instead, they found components of the plant variant of the VCP/p97 complex of yeast and mammals (12, 27) and two putative targets in the form of the transcriptional regulator AGL15 and a Zn-finger protein (12). However, whether or how these are involved in BR signaling remains to be determined. An alternative hypothesis, which is based on the epidermal growth factor (EGF) receptor model, suggests that BRI1-BAK1 dimerization is essential for the activation of both receptor kinases by transphosphorylation (8) and may represent the first bifurcation in the pathway in which both BRI1 and BAK1 participate in downstream signaling.

A final question concerns the role of endocytosis in BR-mediated signaling. Although animal EGF and TGF receptors remain capable of downstream signaling after internalization, so far this has only been shown for the plant receptor FLS2 (28), which is involved in innate immunity and bacterial flagellin perception. BAK1 is reported to enhance endocytosis of BRI1 (11), but whether this also affects their signaling activities is unclear. It is also unknown whether the BRI1-BAK1 complex may have functions that are independent of kinase activity. For example, the kinase activity of another plant receptor, CRINKLY4, is not instrumental for its biological activity (29). If this is a general feature of plant receptors, properties other than the catalytic activity of BRI1 and BAK1 could also be essential for signaling .

In conclusion, it appears that although plant signaling pathways have many components in common with their animal counterparts, their individual properties may be quite different and may help to fully explain how the problem of perceiving and responding to steroid hormones as essential signal molecules has been tackled in different species.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
View Abstract

Navigate This Article