Evolutionarily conserved glycan signal to degrade aberrant brassinosteroid receptors in Arabidopsis

Abstract

Asparagine-linked glycans (N-glycans) are crucial signals for protein folding, quality control, and endoplasmic reticulum (ER)-associated degradation (ERAD) in yeast and mammals. Although similar ERAD processes were reported in plants, little is known about their biochemical mechanisms, especially their relationships with N-glycans. Here, we show that a missense mutation in the Arabidopsis EMS-mutagenized bri1 suppressor 3 (EBS3) gene suppresses a dwarf mutant, bri1-9, the phenotypes of which are caused by ER retention and ERAD of a brassinosteroid receptor, BRASSINOSTEROID-INSENSITIVE 1 (BR1). EBS3 encodes the Arabidopsis ortholog of the yeast asparagine-linked glycosylation 9 (ALG9), which catalyzes the ER luminal addition of two terminal α1,2 mannose (Man) residues in assembling the three-branched N-glycan precursor [glucose(Glc)](3)(Man)(9)[N-acetylglucosamine(GlcNAc)](2). Consistent with recent discoveries revealing the importance of the Glc(3)Man(9)GlcNAc(2) C-branch in generating an ERAD signal, the ebs3-1 mutation prevents the Glc(3)Man(9)GlcNAc(2) assembly and inhibits the ERAD of bri1-9. By contrast, overexpression of EBS4 in ebs3-1 bri1-9, which encodes the Arabidopsis ortholog of the yeast ALG12 catalyzing the ER luminal α1,6 Man addition, adds an α1,6 Man to the truncated N-glycan precursor accumulated in ebs3-1 bri1-9, promotes the bri1-9 ERAD, and neutralizes the ebs3-1 suppressor phenotype. Furthermore, a transfer (T)-DNA insertional alg3-T2 mutation, which causes accumulation of an even smaller N-glycan precursor carrying a different exposed α1,6 Man, promotes the ERAD of bri1-9 and enhances its dwarfism. Taken together, our results strongly suggest that the glycan signal to mark an ERAD client in Arabidopsis is likely conserved to be an α1,6 Man-exposed N-glycan.

Fig. 1.

ebs3-1 mutation inhibits the ERAD of bri1-9. (A) Immunoblot analysis of bri1-9 in ebs3-1 bri1-9. (B) Immunoblot analysis of bri1-9 stability in bri1-9 and ebs3-1 bri1-9. (C) Endo H analysis of bri1-9 in bri1-9 and ebs3-1 bri1-9. For A and C, total proteins from 4-wk-old leaves were treated with or without Endo H, separated by SDS/PAGE, and analyzed by immunoblot with anti-BRI1 antibody. Equal amounts of total proteins were used in A, whereas five times more proteins in bri1-9 than ebs3-1 bri1-9 were used in C, which also contains technical duplicates of Endo H-treated bri1-9 samples. For B, 3-wk-old seedlings were transferred from 1/2 MS-agar medium into liquid 1/2 MS medium containing 180 μM CHX. Equal amounts of seedlings were removed at different time points to extract total proteins into 2× SDS sample buffer, which were subsequently separated by SDS/PAGE and analyzed by immunoblot with anti-BRI1 antibody. The numbers on the left in A and C indicate molecular mass, bri1-9^ER denotes Endo H-sensitive form, and bri1-9^CT represents bri1-9 carrying C-type N-glycans. Dashed lines in A and C show the mobility difference between the bri1-9 band in bri1-9 and that of ebs3-1 bri1-9. Coomassie blue staining of the small subunit of Rubisco (RbcS) serves as the loading control.

Fig. 2.

ebs3-1 is a weak suppressor of the bri1-9 mutant. (A–C) Images of 3-wk-old soil-grown plants (A), 5-d-old etiolated seedlings (B), and 7-wk-old soil-grown mature plants (C) of WT, bri1-9, and ebs3-1 bri1-9. [Scale bars: 1 cm (A and B) and 10 cm (C).] (D) Quantitative analysis of BR sensitivity. Root length of 10-d-old seedlings grown on BL-containing medium were measured and presented as the relative value of average root length of BL-treated seedlings to that of mock-treated seedlings. Each data point represents the average of ∼40 seedlings of duplicated experiments. Error bars denote SE. (E) Immunoblot analysis of BL-induced BES1 dephosphorylation. Total proteins were extracted in 2× SDS buffer from 2-wk-old seedlings treated with or without 1 μM BL for 1 h, separated by SDS/PAGE, and analyzed by immunoblot using an anti-BES1 antibody (27). Star indicates a nonspecific band for loading control.

Fig. 3.

ebs3-1 likely affects assembly of the N-glycan precursor. (A) Immunoblot analysis of BRI1 in WT and ebs3-1 BRI1⁺. (B) Analysis of LLOs in bri1-9, ebs3-1 bri1-9, and a transgenic gEBS4 ebs3-1 bri1-9 line. LLOs of mature plants were extracted, acid-hydrolyzed, fluorescently labeled with PA, and analyzed using SF-HPLC by comparing their elution profiles with that of PA-sugar chain standards (Mn for Man_nGlcNAc₂-PA and G3M9 for Glc₃Man₉GlcNAc₂-PA). Asterisks indicate minor contaminants derived from the labeling process. (C) ebs3-1 affects the electrophoretic mobility of several ER-localized glycoproteins. For A and C, equal amounts of total proteins from 4-wk-old leaves were separated by SDS/PAGE and analyzed by immunoblot with antibodies against BRI1, maize CRT, PDI, or BiP. Numbers on the left indicate molecular mass. Coomassie blue staining of RbcS served as a loading control.

Fig. 4.

EBS3 encodes the Arabidopsis ortholog of yALG9. (A) Images of 3-wk-old transgenic bri1-9 lines carrying an empty vector and transgenic ebs3-1 bri1-9 seedlings carrying a genomic EBS3 transgene (gEBS3) or an empty vector. (Scale bar: 1 cm.) (B) Immunoblot analysis of bri1-9. The numbers on top of the gel panel indicate different transgenic lines. (C) EBS3 complemented the yeast Δalg9 mutation. Growth efficiency of WT or Δalg9 wbp1-2 cells transformed with the vector or an expression plasmid of yALG9, EBS3 or ebs3-1 cDNA. Transformants were spotted in 10-fold serial dilution on synthetic medium and grown for 3 d at 23 °C or 33 °C. (D) Immunoblot analysis of CPY of the WT yeast strain and Δalg9 strain transformed with indicated plasmids. For B and D, equal amounts of total proteins extracted from 4-wk-old soil-grown seedlings (B) or yeast cells of midlog growth phase (D) were separated by SDS/PAGE and analyzed by immunoblot using an anti-BRI1 (B) or anti-CPY (D) antibody. Coomassie blue staining of RbcS served as a loading control in (B). mCPY indicates the mature CPY, CPY*s represent three isoforms of CPY carrying different numbers of truncated N-glycan in (D), and star denotes a cross-reacting band used for loading control.

Fig. 5.

An exposed α1,6 Man residue is likely the glycan signal for bri1-9 ERAD. (A) Schematic structures of suspected major LLOs in ebs3-1 bri1-9, alg3-T2 bri1-9, or transgenic gEBS4 ebs3-1 bri1-9 line. Rectangles denote the two GlcNAc residues and circles represent Man residues with black circles indicating α1,6-linked Man residues. (B) Images of 5-wk-old soil-grown transgenic ebs3-1 bri1-9 mutant containing pPZP212, a gEBS3, or gEBS4 genomic transgene. (C) Immunoblot analysis of the bri1-9 abundance. Numbers in B and C indicate independent transgenic lines. (D) Images of 5-wk-old soil-grown mutants of ebs3-1 bri1-9, bri1-9, and alg3-T2 bri1-9. (E) Immunoblot analysis of PDI and bri1-9. For C and E, equal amounts of total proteins extracted in 2× SDS buffer from 4-wk-old leaves were treated with or without 10 μM Kif, separated by SDS/PAGE, and analyzed by immunoblot with anti-PDI or anti-BRI1 antibody. Coomassie blue staining of RbcS served as a loading control.