Allele-specific suppression of a defective brassinosteroid receptor reveals a physiological role of UGGT in ER quality control

Abstract

UDP-glucose:glycoprotein glucosyltransferase (UGGT) is a presumed folding sensor of protein quality control in the endoplasmic reticulum (ER). Previous biochemical studies with nonphysiological substrates revealed that UGGT can glucosylate nonnative glycoproteins by recognizing subtle folding defects; however, its physiological function remains undefined. Here, we show that mutations in the Arabidopsis EBS1 gene suppressed the growth defects of a brassinosteroid (BR) receptor mutant, bri1-9, in an allele-specific manner by restoring its BR sensitivity. Using a map-based cloning strategy, we discovered that EBS1 encodes the Arabidopsis homolog of UGGT. We demonstrated that bri1-9 is retained in the ER through interactions with several ER chaperones and that ebs1 mutations significantly reduce the stringency of the retention-based ER quality control, allowing export of the structurally imperfect yet biochemically competent bri1-9 to the cell surface for BR perception. Thus, our discovery provides genetic support for a physiological role of UGGT in high-fidelity ER quality control.

Figure 1

ebs1 Suppresses bri1-9 Phenotype by Affecting BR Perception (A) Schematic presentation of the genetic screen for ebs mutants in the bri1-9 background (see “Experimental Procedure” for details) (B) Four-week-old soil-grown plants of bri1-9, wild type and ebs1-1 bri1-9. (C) Seven-week-old mature plants of bri1-9, wild type and ebs1-1 bri1-9 grown in soil. (D) The hypocotyl length of 4-day-old dark-grown seedlings of bri1-9, wild type and ebs1-1 bri1-9. Each bar represents the average measurement of ∼40 seedlings of two duplicate experiments. Error bar denotes standard error. (E) bin2-1/⁺, bin2-1 and their corresponding double mutants with ebs1-1 grown in soil for 6 weeks. (F) det2 and ebs1-1 det2 mutants grown in soil for 6 weeks.

Figure 2

ebs1 Restores BR Sensitivity of bri1-9 (A) Quantitative analysis of BR sensitivity of bri1-9, wild type and ebs1-1 bri1-9 seedlings by the root inhibition assay as described previously (Li et al., 2001). Each data point represents the average root elongation of ∼40 seedlings of two duplicate experiments. Error bar denotes standard error. (B) BR-induced changes in the phosphorylation status of BES1. 3-week-old seedlings of bri1-9, wild type, and ebs1-1 bri1-9 were treated with 1 μM BL for 1 hr in liquid half MS medium. Total proteins were extracted with 2X SDS buffer and separated by a 10% SDS-PAGE. BES1 was detected by anti-BES1 antibody (Mora-Garcia et al., 2004). Non-specific bands (*) were used as a loading control in the lower panel. (C) Feedback inhibition of CPD expression by BL. Total mRNAs were extracted from BL-treated seedlings and probed for CPD as described previously (Li et al., 2001). The filter was stripped and reprobed with the 18S rDNA for loading control.

Figure 3

ebs1 is an Allele Specific Suppressor of the bri1-9 Mutation (A) ebs1 cannot suppress bri1-301 and bri1-101 mutants containing a kinase domain mutation. From left to right: 6-week-old soil-grown plants of bri1-301, ebs1-1 bri1-301, bri1-101 and ebs1-3 bri1-101. (B) ebs1 fails to suppress bri1-6 and bri1-113 mutants harboring mutations in the BR-binding domain. From left to right: 5-week-old soil-grown plants of bri1-6, ebs1-1 bri1-6, bri1-113, and ebs1-1 bri1-113.

Figure 4

Molecular Characterization of the EBS1 Gene (A) Genetic mapping of EBS1. EBS1 was mapped to a 63-kb region at the bottom of chromosome I between markers CER469884 and F3I17_2, which is covered by two overlapping BAC clones: F23N20 and F3I17. Molecular makers and numbers of recombination for each marker are shown above and below the line, respectively. Sequence analysis of the entire 63-kb region of ebs1-1 identified a single nucleotide mutation in At1g71220 composed of 37 exons (bar) and 36 introns (line). Arrows indicate the positions of 5 ebs1 mutations. (B) Summary of 5 ebs1 alleles, their predicted molecular defects and the corresponding positions in the human UGGT (accession # Q9NYU2).

Figure 5

bri1-9 is Retained in the ER but Moves to the PM in ebs1 Mutants (A) Endo H sensitivity of BRI1 and bri1-9. Total proteins from bri1-9, wild type, ebs1-1 bri1-9 to ebs1-5 bri1-9 were subjected to Endo H treatment followed by immunoblotting with anti-BRI1 antiserum. (B) to (D) Western blotting analysis of membrane fractions obtained through aqueous two phase partitioning of total microsomal proteins isolated from bri1-9 (B), wild type (C) and ebs1-3 bri1-9 (D) plants using antibodies against BRI1, BiP, PDI, PMA2, PIP2 and. M, U, and L indicate total microsomal fraction, upper phase that enriches PM proteins, and the lower phase that was depleted of PM, respectively. (E) to (M) Confocal analysis of subcellular localization of BRI1:GFP and bri1-9:GFP in root tips of 6-day-old light-grown seedlings coexpressing DsRed:HDEL and BRI1:GFP (E-G) or bri1-9:GFP in either EBS1⁺ (H-J) or ebs1 (K-M) background.

Figure 6

The ebs1 Mutation Inhibits Monoglucosylation of bri1-9 (A) Schematic illustration of the JBαM action. Hexagon, circle, and diamond represent glucose, mannose, and N-acetylglucosamine residues, respectively. (B) Sensitivity of BRI1:GFP and bri1-9:GFP to JBαM treatment. Anti-GFP immunoprecipitates of BRI1:GFP, bri1-9:GFP, and bri1-9:GFP ebs1-3 transgenic plants were incubated with or without JBαM followed by Western blotting analysis with an anti-GFP antibody.

Figure 7

bri1-9 Interacts with CNXs and BiPs Coimmunoprecipitation of bri1-9:GFP with BiPs and CNXs. Total protein crude extracts from wild type plants (lane 1), or transgenic plants of BRI1:GFP (lane 2), bri1-9:GFP (lane 3), or bri1-9:GFP ebs1-3 (lane 4) were immunoprecipitated with anti-GFP antibody and analyzed by Western blotting with antibodies against GFP, a maize CRT, or BiP. The left and right 4 lanes show the presence of BRI1:GFP or bri1-9:GFP, CNXs, CRTs, and BiPs in total protein extracts and anti-GFP immunoprecipitates, respectively. The numbers represent the relative signal intensity of co-immunoprecipitated CNXs or BiPs after normalization against the anti-GFP signal.