GSK3s and brassinosteroid signalling in flowering plants
Steroid hormone signalling is found in
plants, animals and fungi and, therefore, might be predicted to have an ancient
evolutionary origin [12]. However, the proteins involved in plant steroid signalling belong
to plant-specific protein families, thus indicating a separate origin for
steroid signalling in the plant lineage. Brassinosteroid (BR) signalling
modulates a range of physiological responses in flowering plants, including
cell expansion, greening, flowering time, fertility and differentiation of
vascular tissue (reviewed in [13-16]).
In Arabidopsis, BRs are perceived by a plasma membrane-localized
receptor kinase, BRI1 [13-15]. The
downstream signal transduction involves a gene isolated by forward genetic
studies, BRASSINOSTEROID-INSENSITIVE 2 (BIN2). Dominant mutations in bin2 result in brassinosteroid-insensitive
dwarf plants. BIN2 encodes a GSK3
with a catalytic domain sharing ~70% identity with animal GSK3β [17-20]
(Figure 1). Gain-of-function bin2
point mutations and over-expression of BIN2 all cause the same BR insensitivity
phenotype, confirming BIN2 as an inhibitor of BR signalling [17-20].
In the absence of a BR
signal, active BIN2 negatively regulates BR-specific transcription factors,
BRASSINAZOLE RESISTANT 1 (BZR1) and bri1-EMS-SUPPRESSOR
1 (BES1/BZR2) (Figure 2). BIN2 phosphorylates BZR1/BES1 on many serine and
threonine residues, which leads to their proteasomal degradation [21-23],
reminiscent of the role of animal GSK3 during both Wnt- and Hedgehog signalling
[4, 24]. BIN2
can also promote BZR1/BES1 nuclear export, thus inhibiting BZR1/BES1 binding to
DNA [25-26]. Again,
this is a role also demonstrated by activated (phosphorylated (Figure 1))
animal and amoebal GSK3, which control the nuclear export of the transcription
factors NF-ATc and StatA, respectively [27-28]. Interaction
between BIN2 and BZR1/BES1 is via a specific docking mechanism not seen in
other GSK3 substrates [29], and independent
of prime phosphorylation or a scaffold protein [21]. In
the presence of BR, BIN2 activity towards BES1/BZR1 is inhibited and,
consequently, unphosphorylated BZR1 and BES1 are no longer degraded and can
direct transcription of BR-responsive genes (Figure 2) [13-15].
The exact mechanism by
which the BR signal is transduced from the plasma membrane to BIN2 has now been
elucidated [13-15,30-32] and
it demonstrates novel modes of inhibition of GSK3 activity. BR is perceived by
the BRI1 brassinosteroid receptor and BAK1 (BRI1-associated receptor kinase) co-receptor
(Figure 2). BR binding to BRI1 enables transphosphorylation of BRI1 and BAK1,
and activation of BRI1. BRI1 then phosphorylates BR-signalling kinases (BSKs)
at the plasma membrane. Phosphorylated BSK binds and activates BSU1
phosphatases, which dephosphorylate tyrosine 200 (Y200) in BIN2, inactivating
its kinase activity [31]
(Figure 2). Interestingly, the dominant bin2-1
(E263K) mutation (Figure 1) renders BIN2 hyperactive by blocking its
dephosphorylation by BSU1 [31]. In
response to BR, BIN2 is also degraded by the proteasome, whereas the bin2-1 mutant protein is considerably
stabilized [30].
Whether it is the Y200-dephosphorylated form of BIN2 that is specifically
targeted for degradation has yet to be determined. Y200 in BIN2 is homologous
to Y216 in human GSK3β (Figure 1), a tyrosine residue absolutely required for
GSK3 activity. The homologous tyrosine residue in other AtGSK3s is also phosphorylated in
vivo, and this phosphorylation is required for kinase activity [10]. In
GSK3β, this residue is autophosphorylated by GSK3 itself just after the protein
is synthesized, with the help of the Hsp90 chaperone protein [33].
BSU1-mediated dephosphorylation of Y200 in BIN2 is the first example of direct
dephosphorylation of the conserved tyrosine residue of GSK3 as a means of
negatively regulating kinase activity. Furthermore, most plant GSK3s (all except
clade III kinases, Table 1) do not contain the N-terminal serine 9 residue
found in GSK3β that is key to regulating GSK3β inhibition (Figure 1) (reviewed
in [2-3]). It
is, therefore, possible that a different mechanism exists in plants for GSK3
inactivation.
The mechanism for
brassinosteroid perception and transduction involves GSK3 homologues
functioning redundantly for the negative regulation of BR responses [26,34-35].
Experiments using loss- and gain-of-function mutants showed that two other GSK3
homologues, AtSK22/BIL1 and AtSK23/BIL2 (Table 1), function in the
same way as BIN2 during BR signalling [35].
Together these genes make up sub-clade II of the Arabidopsis GSK3s in the flowering plant phylogeny [9,36]
(Table 1). Although the triple mutant lines of BIN2, BIL1 and BIL2 show a BR-enhanced
phenotype, analysis of the levels of phosphorylated BZR1 showed that despite
knockouts of all clade II GSK3s there was still substantial phosphorylation of
BZR1 [35]. This
indicated a role for additional GSK3(s) in the BR signalling pathway, which
have been identified as AtSK31/ASKθ [37], a member of sub-clade III, and possibly all three of the subclade
I GSK3s [31].
Therefore, at least seven out of the ten Arabidopsis
GSK3 proteins are implicated in BR signalling. Importantly, a chemical activator
of BR signalling, bikinin, specifically inhibits the same subset of seven Arabidopsis
GSK3 proteins [38]. BIN2
orthologues from cotton (Gossypium hirsutum) and rice (Oryza sativa) have been isolated and their overexpression in Arabidopsis has been shown to cause severe
growth defects similar to bin2
gain-of-function (BR-insensitive) mutants [39-40]. This
suggests that BIN2 orthologues encode isoforms that share a conserved function
in BR signalling.
Post Comment
No comments