Another tool in the receptor endocytosis kit: ligand-insensitive BRASSINOSTEROID INSENSITIVE1

Brassinosteroids (BRs) are growth-promoting hormones that play numerous roles throughout the plant body and life cycle. Arabidopsis (Arabidopsis thaliana) plants that lack the ability to produce or respond to BRs display a characteristic set of phenotypes: dwarfed stature, dark green leaves, male sterility, and photomorphogenic development in the dark. The biologically most active form of BR, called brassinolide (BL), is recognized by the leucine-rich repeat receptor-like kinase (LRR-RLK) BRASSINOSTEROID INSENSITIVE1 (BRI1) (Clouse et al. 1996; Li and Chory 1997). BL binds to a ∼70 amino acid “island domain” between LRR repeats 21 and 22 on the extracellular side (Hothorn et al. 2011). This binding sets in motion a complex series of molecular events, including recruitment of the coreceptor BRI1-ASSOCIATED KINASE1 (BAK1) and both phosphorylation and ubiquitylation of the cytoplasmic domain, which leads to activation of the BR signaling pathway. The ensuing signaling cascade culminates with the dephosphorylation and activation of the transcription factor BRI1-EMS-SUPPRESSOR1 (BES1) (Kim and Wang 2010).

BRI1 cycles between the plasma membrane and early endosomes (Russinova et al. 2004; Geldner et al. 2007), but the extent to which ligand binding and post-translational modification contribute to BRI1 endocytosis is still unclear. Early studies concluded that BRI1 cycling is constitutive and independent of ligand binding (Geldner et al. 2007). However, ligand-dependent ubiquitylation was later shown to promote BRI1 endocytosis (Zhou et al. 2018). In this issue of Plant Physiology, Claus and collegues (Claus et al. 2023) developed another version of BRI1, called BRI1^Q, as a tool to address how ligand perception affects the trafficking of BRI1.

To generate BRI1^Q, the authors first inspected the structure of BRI1 bound to BL and identified five key residues that contribute to the high-affinity binding. These residues were mutated either to alanine or to the corresponding residue found in BRI1-LIKE2 (BRL2), a homolog that does not bind BL (Fig. 1A). The resulting BRI1^Q completely lacked BL-binding activity in vitro. When expressed in plants, BRI1^Q accumulated to a similar level as the wild-type BRI1 but did not interact with BAK1 and did not show the ligand-induced phosphorylation and ubiquitylation modifications observed in the wild-type. BRI1^Q also did not support the dephosphorylation of the downstream transcription factor BES1, consistent with the in vitro result. Consequently, BRI1^Q failed to complement a bri1 null mutant, even though the mutant protein localized correctly in the cell (Fig. 1B).

Figure 1.

Structure-guided engineering of BRI1. A) Five amino acid changes (Y597M/Y599F/Y642A/M657E/F681A) in the LRR-domain of BRI1 generated BRI1^Q, which does not bind BL. B) The nonbinding BRI1^Q allele fails to complement the bri1 null mutant (top, scale bar 2 cm) but localizes correctly in the cells, similar to the wild-type BRI1 (bottom, scale bar 10 µm).

With BRI1^Q as a control, the authors then set out to test the importance of ligand binding for BRI1 internalization and trafficking. To do this, they tagged both BRI1 and BRI1^Q with a fluorescent protein and measured how quickly the plasma membrane-bound receptor entered the cytosol. In one set of experiments, they added the trafficking inhibitor brefeldin A (BFA), which causes aggregation of early endosomes containing internalized BRI1/BRI1^Q. They could not detect any significant difference in the rate of endocytosis or accumulation in BFA aggregates, indicating that the ability to interact with BL is not critical to BRI1 endocytosis. They obtained the same result when BRI1^Q was expressed in the bri1 mutant and the wild-type, eliminating the possibility that BRI1^Q somehow associates with the wild-type protein to “hitch a ride” into the cell. Essentially the same result was obtained using an inducible expression system to generate a pulse of BRI1/BRI1^Q. In this case, both proteins were first seen accumulating at the plasma membrane, after which they were internalized with nearly indistinguishable kinetics. Taken together, these results indicate that ligand binding does not play a major role in BRI1 endocytosis. However, the experiments were done with endogenous (unknown) concentration of BL, and it is possible that an experimentally induced switch between a low and a high BL state (e.g. by pretreatment with a biosynthesis inhibitor followed by BL re-application) might reveal a more subtle differences between the wild-type receptor and the nonbinding BRI1^Q.

The BRI1^Q allele also provides a tool to study the BR signaling-independent functions of BRI1. Plants that lack BRI1 have more metaxylem cells in the developing root than in BR biosynthesis mutants, probably because BRI1 sequesters the receptor-like protein RLP44 (Holzwart et al. 2020). BRI1^Q retained the ability to interact with RLP44 and partially complemented the ectopic metaxylem differentiation phenotype of bri1, supporting this BR-independent role of BRI1 in root development.

BR signaling is a well-understood signaling pathway in plants and has become a test system to investigate how ligand perception and intracellular trafficking (endocytosis, recycling, and vacuolar degradation) are connected. The work by Claus et al. provides an excellent tool to probe key features of this system, which may impact our understanding of signaling in the large family of plant receptor-like kinases (De Smet et al. 2009).