Advanced proteomic analyses yield a deep catalog of ubiquitylation targets in Arabidopsis

Abstract

The posttranslational addition of ubiquitin (Ub) profoundly controls the half-life, interactions, and/or trafficking of numerous intracellular proteins. Using stringent two-step affinity methods to purify Ub-protein conjugates followed by high-sensitivity mass spectrometry, we identified almost 950 ubiquitylation substrates in whole Arabidopsis thaliana seedlings. The list includes key factors regulating a wide range of biological processes, including metabolism, cellular transport, signal transduction, transcription, RNA biology, translation, and proteolysis. The ubiquitylation state of more than half of the targets increased after treating seedlings with the proteasome inhibitor MG132 (carbobenzoxy-Leu-Leu-Leu-al), strongly suggesting that Ub addition commits many to degradation by the 26S proteasome. Ub-attachment sites were resolved for a number of targets, including six of the seven Lys residues on Ub itself with a Lys-48>Lys-63>Lys-11>>>Lys-33/Lys-29/Lys-6 preference. However, little sequence consensus was detected among conjugation sites, indicating that the local environment has little influence on global ubiquitylation. Intriguingly, the level of Lys-11-linked Ub polymers increased substantially upon MG132 treatment, revealing that they might be important signals for proteasomal breakdown. Taken together, this proteomic analysis illustrates the breadth of plant processes affected by ubiquitylation and provides a deep data set of individual targets from which to explore the roles of Ub in various physiological and developmental pathways.

Figure 1.

Use of Tandem Arrays of Ub-Binding Domains to Enrich for Ubiquitylated Proteins. (A) Schematic representation of the GST-tagged versions of 2xUSU and TUBEs proteins generated from the USU region of human HHR23A. (B) SDS-PAGE analysis of purified GST and the GST-tagged versions of USU, 2xUSU, and TUBEs. The calculated molecular masses are indicated. (C) Poly-Ub chain binding in vitro. Mixtures of poly-Ub chains linked via either Lys-11 (K11), Lys-48 (K48), or Lys-63 (K63) were incubated with recombinant GST or GST fused to USU, 2xUSU, or TUBEs and precipitated with glutathione binding beads. The Ub moieties were resolved by SDS-PAGE and detected by immunoblot analysis with anti-Ub antibodies. The arrowheads and brackets to the right identify the various Ub polymers. Full strength (5 μg) and a fivefold dilution (1 μg) of each poly-Ub chain preparation are shown on the left. (D) Quantification of poly-Ub chain binding efficiency for each Ub-binding domain by densitometric scanning of the immunoblots shown in (C). Each bar represents the average (±sd) of three independent experiments. (E) Flow chart describing the proteomic analysis of Ub conjugates isolated by two-step affinity protocols from intact Arabidopsis seedlings expressing 6His-Ub. The first purification step uses Ub affinity under native conditions followed by a second purification step that uses Ni-NTA affinity under denaturing conditions. ESI, electrospray ionization.

Figure 2.

Two-Step Affinity Purification of Ubiquitylated Proteins from Transgenic Arabidopsis Expressing 6His-Ub. Ub conjugates were purified from hexa(6His-UBQ) seedlings by the two-step protocol outlined in Figure 1E and subjected to SDS-PAGE. The gels were either stained for total protein with silver or immunoblotted with anti-Ub or anti-5His antibodies. Samples from wild-type seedlings (WT) subjected to the same purification with GST and Ni-NTA beads were included for comparison. FT and EL represent the flow-through and elution, respectively, from the Ub affinity (1) and Ni-NTA affinity columns (2). “Crude” represents the clarified crude extracts before purification. The migration position of free Ub, the Ub dimer, and Ub conjugates are indicated on the right of the anti-Ub antibody blot.

Figure 3.

Venn Diagrams Showing the Overlap of Ubiquitylated Arabidopsis Proteins Identified by MS from Samples Purified using 2xUSU or TUBEs. (A) and (B) Overlap of ubiquitylated proteins isolated using 2xUSU or TUBEs with or without pretreatment of the Arabidopsis seedlings with MG132. (C) Overlap of ubiquitylated proteins isolated using 2xUSU and TUBEs. (D) Overlap of the complete list of ubiquitylated proteins purified via the 2xUSU and TUBEs matrices from seedlings with or without pretreatment with MG132.

Figure 4.

Sorting of Ubiquitylated Arabidopsis Proteins into Functional Categories. The percentages of proteins in each data set were classified into functional categories using the GO annotations in the MIPS-FunCat database. (A) The complete Arabidopsis proteome of 27,430 proteins. (B) The 1013 proteins identified by MS analysis of crude extracts from green wild-type seedlings prior to enrichment. (C) The combined list of 941 ubiquitylated proteins isolated from hexa(6His-UBQ) seedlings using either 2xUSU or TUBEs followed by Ni-NTA affinity chromatography. (D) Those proteins in (C) that became more abundant or were solely detected after MG132 treatment of seedlings. (E) Those proteins in (C) whose levels were unaffected or reduced after MG132 treatment of seedlings. (F) Overlap of Arabidopsis proteins from each functional category based on their MS detection in crude seedling extracts (CE) and in the MS data sets of affinity-purified ubiquitylated proteins obtained from seedling exposed to DMSO (−MG132) or MG132 (+MG132). The numbers associated with each space represents the total number of MS-detected proteins for that space. The number at the top of each Venn diagram represents the total number of proteins in each category for the complete Arabidopsis proteome.

Figure 5.

Protein Interaction Networks of Identified Ubiquitylated Proteins. Protein interaction networks were generated with the complete list of ubiquitylated proteins using the Search Tool for the Retrieval of Interacting Genes/Proteins database and visualized using the Cytoscape program. The networks shown were restricted to those with at least one interaction. The node colors reflect ubiquitylation states that increased (red) or were unaffected or reduced (stable/reduced; blue) upon MG132 treatment based on PSMs. Important functional clusters are surrounded by dashed lines. Blue asterisk locates nitrate reductase. Either the Arabidopsis locus identifier or the protein name abbreviation (where known) is shown for each substrate.

Figure 6.

The Abundance of Proteasome- and Ribosome-Associated Proteins Are Inversely Affected by MG132. (A) Fold change in ubiquitylated protein abundance based on MS PSMs after a 12-h pretreatment of seedlings with 50 μM MG132. Each bar represents the average of three independent experiments (±sd). Ubiquitylated proteins were purified by the 2xUSU or TUBEs matrices from hexa(6His-UBQ) seedlings. Increased category includes 100 ubiquitylated proteins with a greater than or equal to threefold increase in PSMs. Unaffected/reduced category includes 266 ubiquitylated proteins with less than threefold change in PSMs. Proteasome and ribosome categories include 20 and 36 ubiquitylated proteins, respectively. Contam category represents 50 proteins that were purified from wild-type seedlings by sequentially GST and Ni-NTA chromatographic steps. (B) Levels of proteasome-associated proteins increase, whereas ribosome proteins decrease in abundance upon MG132 treatment. Crude extract proteins from 10-d-old seedlings [the wild type (WT) and hexa(6His-UBQ)] with or without MG132 treatment were separated by SDS-PAGE and subjected to immunoblot analysis with antibodies that recognize the proteasome subunits PBA1, RPN1, RPN5, RPT2, and RPT4, the proteasome accessory factors PA200 and CDC48, and the ribosome 40S particle subunits S3 (RPS3) and S5 (RPS5). Anti-histone H3 antibodies were used to confirm equal protein loading.

Figure 7.

The Distribution of Intra-Ub–Ub Linkages and Their Changes in Abundance upon Treating Arabidopsis with MG132. (A) Three-dimensional ribbon front and side diagrams of plant Ub highlighting the Lys residues that could bind Ub covalently. α-Helices, β-strands, and the side chains of Lys residues are shown in cyan, green, and red, respectively. N, N terminus; C, C terminus; G76, C-terminal active-site Gly. Adapted from Protein Data Bank code 1UBQ (Vijay-Kumar et al., 1987). (B) and (C) Percentage of Ub footprints identified by MS/MS that represent Ub chains linked internally through the various Ub Lys residues. Ubiquitylated proteins were purified with TUBEs (B) or 2xUSU (C) from seedlings with or without pretreatment with 50 μM MG132. Each bar represents the average of three independent MS analyses (±sd). Asterisks indicated significant differences as determined by the Student’s t test (P < 0.05).