Identification of SUMO conjugation sites in the budding yeast proteome

Abstract

Post-translational modification by the small ubiquitin-like modifier (SUMO) is an important mechanism regulating protein function. Identification of SUMO conjugation sites on substrates is a challenging task. Here we employed a proteomic method to map SUMO acceptor lysines in budding yeast proteins. We report the identification of 257 lysine residues where SUMO is potentially attached. Amongst the hits, we identified already known SUMO substrates and sites, confirming the success of the approach. In addition, we tested several of the novel substrates using SUMO immunoprecipitation analysis and confirmed that the SUMO acceptor lysines identified in these proteins are indeed bona fide SUMOylation sites. We believe that the collection of SUMO sites presented here is an important resource for future functional studies of SUMOylation in yeast.

Keywords: SUMO; budding yeast; mass spectrometry; proteome; site-specific SUMOylation.

Figure 1. FIGURE 1: Proteomic screen to identify SUMO sites in budding yeast proteins.

(A) Sequences of wildtype SUMO (SMT3) and the lysine-deficient His6 tagged mutant (SMT3-KallR I93R) used in the method. (B) Wild-type and SMT3-KallR I93R strains plated on full media (YPD) and media containing methyl methanesulphonate (MMS). (C) Diagrammatic representation of the purification strategy employed to enrich for SMT3-conjugated peptides. Cell lysates from yeast expressing SMT3-KallR I93R were digested with the endoprotease LysC, cleaving after lysines. The digested lysates were run on SDS-PAGE. Unconjugated SMT3-KallR I93R (indicated) and SMT3-KallR I93R conjugated to target protein fragments were identified. Gel area above the unconjugated SMT3-KallR I93R band were excised, digested with trypsin, and analyzed by nano LC-MS/MS. And database searched to identify SUMO acceptor lysines in the yeast proteome.

Figure 2. FIGURE 2: SUMO-acceptor lysines detected in our mass spectrometry analysis were compared with SUMO-acceptor lysines already described in previous studies done in S. cerevisiae.

(A) Previous studies were based on either MS or site-directed mutagenesis/immunoblotting, or a combination of both. Most of the SUMO-acceptor lysines previously found were also detected in this study. SUMO substrates (total of 257) identified in the yeast proteome of cells grown asynchronously. Previously published SUMO-substrates were obtained . (B) To ensure that diglycine-modified lysines detected by mass spectrometry after Smt3 purification are not due to modification by ubiquitin or Nedd8, a centromeric plasmid 8His-SMT3-KallR-REQIGG-pRS415 expressing the Smt3 variant used previously for the Smt3 purification protocol with the difference that the RGG conjugating terminus was replace for the native RIEQGG C-terminus was employed. SUMO-acceptor lysines modified by the 8His-Smt3-KallR-REQIGG keep a side chain of 5 aa after trypsin digestion (EQIGG). Therefore, any diglycine-modified lysines detected by mass spectrometry under these conditions can only be due to either false positive hits, or to ubiquitinated or neddylated contaminants. A large culture of 9 l of the strain expressing the 8His-SMT3-KallR-REQIGG variant was grown in YPD and harvested at O.D. 0.9. Smt3 purification and mass spectrometry analysis was performed as described in Material and Methods. We detected 23 diglycine-modified lysines. None of these corresponded with previously detected diglycine-modified lysines in our Smt3-RGG pulldowns. In addition none of these diglycine-modified lysines are within SUMO consensus sequences. This strongly indicates that sumo-acceptor lysines identified after purification of Smt3-RGG pulldowns represent bona fide SUMOylation sites.

Figure 3. FIGURE 3: Mutational analysis of SUMO sites identified.

(A) Histidine pulldowns from cells carrying TFG1, RAD16 and FOB1 tagged with 3 HA epitopes in strains expressing SMT3 tagged with 6 histidines (6his-SMT3) or wild-type SMT3 (untagged control). Western blot analysis using α-HA antibodies in the pulldown demonstrates the presence of SUMO conjugates for TFG1, RAD16 and FOB1. (B) Histidine pulldowns from cells carrying RPC53 or RPC53 (K38,115,237,322R) tagged with 3 HA epitopes. The analysis demonstrates that lysines 38, 115, 237 and 322 (identified in our screen) are sites for the conjugation of SMT3 in RPC53. (C) Histidine pulldowns from cells carrying NTG1 or NTG1 (K38, 396R) tagged with 3HA epitopes. The analysis demonstrates that lysines 38 and 396 (identified in our screen) are sites for the conjugation of SMT3 in NTG1.