The brassinosteroids (BRs) are a group of compounds closely related to the mammalian steroid hormones. While the existence of BRs had been known for some years, their functions in growth and development have only become clear following molecular genetics studies in Arabidopsis. Mutants in BR biosynthesis suggest that this hormone has important roles in cell elongation, xylem differentiation and photomorphogenesis.
The BR biosynthesis pathway is a continuation of the sterol biosynthesis pathway: the early committed steps to the sterol campesterol are catalysed by enzymes with similarity to mammalian steroid reductases. The pathway from campesterol to the active BRs, castasterone and brassinolide (Fig. 4), is best viewed as a network with the enzymes involved being capable of catalysing conversions of multiple substrates. All the biosynthetic enzymes which have been characterised at the molecular level in this part of the pathway to date are cytochrome P450s.
The P450 CYP90B1 from Arabidopsis was identified by the BR-responsive dwf4 (dwarf 4) mutant (Choi et al. 1998). BR biosynthetic intermediates were fed to the mutant to define the block in the BR pathway. The mutant phenotype was rescued with the 22a-hydroxylated intermediates 6-de-oxocathasterone and cathasterone (Fig. 4), as well as other 22a-hydroxyl-
ated intermediates (not shown in Fig. 4), and accumulated non-22a-hydroxyl-ated intermediates (including campestanol and 6-oxocampestanol), strongly suggesting that CYP90B is capable of carrying out the 22a-hydroxylation of multiple substrates.
The 23a-hydroxylation steps in BR biosynthesis are catalysed by CYP90A1 in Arabidopsis, a P450 with 43% amino acid sequence identity to CYP90B1. This P450 was also identified by a BR biosynthesis mutant, cpd (constitutive photomorphogenesis and dwarfism; Szkeres et al. 1996). Feeding studies to rescue the mutant phenotype were used to define the likely reactions catalysed by this P450. Application of 6-deoxoteasterone or teasterone rescued the cpd mutant phenotype, whereas 6-deoxocathasterone and cathasterone did not, suggesting that CYP90A1 catalyses the 23a-hydroxylation steps in the BR biosynthesis pathway.
A rice dwarf mutant, d2, has a mutation in the gene encoding CYP90D2; this mutant accumulates 6-deoxocathasterone and the mutant phenotypes are rescued by 3-dehydro-6-deoxoteasterone and 3-dehydroteasterone but not by earlier intermediates (Hong et al. 2003), suggesting it is blocked between these intermediates. Another rice mutant in a P450 related to the CYP90 family of P450s, CYP724B1, has phenotypes including dwarfism that can be rescued by BR application. Feeding of intermediates suggests that the block in this mutant is in the steps of the pathway from 3-dehydro-6-deoxoteasterone and 3-dehydroteasterone to 6-deoxotyphasterol and typhasterol, respectively (Tanabe et al. 2005). So far there are no homologues of the CYP724B1 subfamily reported in other species.
CYP92A6 (DDWF1; dark-induced DWF-like protein 1) was isolated from pea in a yeast two-hybrid screen with the G-protein Pra2 (see below; Kang et al. 2001). This P450 has been shown by green fluorescent protein (GFP) fusion to be localised to the ER. Analysis of BR biosynthesis intermediates and feeding studies in plants carrying an antisense to CYP92A6 suggest that CYP92A6 catalyses the C-2 hydroxylations of 6-deoxotyphasterol and typhasterol to 6-deoxocastasterone and castasterone, respectively. This was confirmed by assaying bacterially expressed CYP92A6 mixed with yeast microsomes and demonstrating that the mix carried out the C-2 hydroxy-lation of typhasterol to castasterone. In Arabidopsis there is no CYP92A6 homologue, however evidence from quantifying intermediates in the BR biosynthesis pathway in a CYP90C1 mutant suggests that it catalyses the C-2 hydroxylations of 6-deoxotyphasterol and typhasterol, as both these intermediates accumulate in the mutant and the concentrations of the products 6-deoxocastasterone and castasterone are reduced (Kim et al. 2005). The CYP90C1 mutant could only be rescued by application of castasterone or brassinolide, which provides supporting evidence that it is a C2-hydroxylase. The CYP90C1 mutant (also known as rotundifolia3) does not show the classical BR-deficient dwarf phenotype but does have reduced petiole elongation and more rounded leaves. When combined with a mutant in the closely re lated CYP90D1, which has no phenotype as a single mutant in Arabidopsis, a plant with a strong BR-deficient phenotype is produced, suggesting there is some degree of functional redundancy between these two P450s. At present there is no evidence for the Arabidopsis CYP90D1 acting at the earlier steps in the pathway where the rice CYP90D2 is likely to act.
CYP85A1 was isolated using the tomato dwarf mutant and has been characterised by expression of the CYP85A1 cDNA in yeast together with a cytochrome P450 reductase (Bishop et al. 1999; Shimada et al. 2001). This analysis demonstrated that this P450 carries out the C-6 oxidation of multiple substrates (9-deoxoteasterone, 3-dehydro-6-deoxoteasterone, 6-deoxotyphasterol and 6-deoxocastasterone) in a reaction with two distinct steps. An activation tagging strategy in Arabidopsis identified CYP72B1 as a potential brassi-nosteroid catabolic enzyme. In the activation-tagged plants the CYP72B1 was overexpressed, the plants had low endogenous brassinolide and accumulated 26-hydroxybrassinolide (Neff et al. 1999). Subsequent expression of the CYP72B1 cDNA in yeast has shown that it catalyses the C-26 hy-droxylation of both castasterone and brassinolide to 26-hydroxycastasterone and 26-hydroxybrassinolide, respectively (Turk et al. 2003). Castasterone and brassinolide both have biological activity, whereas 26-hydroxycastasterone and 26-hydroxybrassinolide are biologically inactive.
It is therefore likely that BR biosynthesis from campestanol and 6-oxo-campestanol is exclusively catalysed by cytochrome P450s. The lactonisation of the B ring to form brassinolide from castasterone is the one step for which an enzyme has not been identified in any system, but this is also likely to be a P450-catalysed reaction. The P450s involved in BR biosynthesis are quite closely related, with the CYP90 family playing a prominent role. The evidence from some of the weaker mutant phenotypes is that there may be some functional redundancy, with multiple enzymes catalysing the same reaction. This could be resolved by heterologous expression of the enzymes; however, to date it has not been possible to assay the activities of any of the CYP90 P450 family by expression in yeast. The P450 enzymes of the BR biosynthesis pathway all appear capable of carrying out reactions at the same position on the BR skeleton on different substrate molecules, thus allowing BR biosynthesis to proceed as a network rather than a linear pathway. As the location of all the P450 enzymes in the BR pathway, the ER plays a central role in the biosynthesis and catabolism of this plant hormone.
Was this article helpful?