N-Terminal Modifications of Ubiquitin via Methionine Excision, Deamination, and Arginylation Expand the Ubiquitin Code

Purification of Ub-GST fusions and their derivatives

pCH6913 (expressing ^6HisUb-^MQUb_AA-GST), pCH6914 (expressing ^6HisUb-^QUb_AA-GST), pCH6915 (expressing ^6HisUb-^EUb_AA-GST), pCH6916 (expressing ^6HisUb-^REUb_AA-GST), pCH7047 (expressing ^6HisUb-MQIFVKTLTGK-GST), pCH7048 (expressing ^6HisUb-QIFVKTLTGK-GST), pCH7049 (expressing ^6HisUb-EIFVKTLTGK-GST), and pCH7050 (expressing ^6HisUb-REIFVKTLTGK-GST) were transformed into BL21 (DE3) Escherichia coli. Five milliliters of overnight culture of these transformants was inoculated into 500 ml of Lysogeny Broth (LB) medium with ampicillin (final concentration of 100 _µg/ml), followed by growth at 37°C to A600 of ~0.6. Expression of the Ub fusions was induced by incubating E. coli cells with 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) (Cat. No. I2481C; GoldBio, USA) at 18°C overnight.

Cell pellets were thawed and resuspended in lysis buffer (50 mM HEPES, 150 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride [PMSF, Cat. No. 10837091001; Sigma-Aldrich, USA], 1 mM DTT, 0.1% Triton X-100) with lysozyme (final concentration of 1 mg/ml). Resuspended cells were incubated on ice for 20 min and disrupted by sonication (VCX-130; SONICS, USA), five times for 30 s each, at 1 min intervals with 40% amplitude. Cell lysates were centrifuged at 25,000 × g for 20 min at 4°C and the supernatants were incubated with 0.5 ml of pre-washed Glutathione Sepharose 4B (Cat. No. GE17075605; GE Healthcare, USA) at 4°C for 4 h. After washing the beads three times in 10 ml of washing buffer (50 mM HEPES, 150 mM NaCl, 10% glycerol, 0.1% Triton X-100, 1 mM DTT), the bound proteins were eluted with 6 ml of elution buffer (50 mM HEPES, 150 mM NaCl, 10% glycerol, 0.1% Triton X-100, 1 mM DTT, 30 mM reduced glutathione, pH 8.0), followed by 4 h of dialysis against storage buffer (25 mM Tris-HCl pH 7.4, 150 mM NaCl, 10% glycerol, 1 mM DTT) and overnight incubation with the catalytic core Usp2 (Usp2-cc) at 4°C. The Usp2-cc treatment yielded free ^6HisUb and MQIFVKTLTGK-GST, QIFVKTLTGK-GST, EIFVKTLTGK-GST, REIFVKTLTGK-GST, ^MQUb_AA-GST, ^QUb_AA-GST, ^EUb_AA-GST, or ^REUb_AA-GST. To remove the free ^6HisUb molecules, the Usp2-cc-treated samples were incubated with Ni-NTA resins. Subsequently, the Ni-NTA bead-unbound proteins were re-dialyzed against the above storage buffer and further purified by gel filtration chromatography. The active fractions were collected, concentrated, and stored at −80°C.

Production of anti-^REUb

The synthetic peptide REIFVKTLTGKC was used for the production of anti-^REUb as the antigen. Rabbit polyclonal antisera to the peptide were raised by AbClon (Korea). Antibodies specific for REIFVKTLTGKC were “negatively” selected from antiserum by sequentially incubating 5 ml of clarified antiserum with 1 ml of Glutathione Sepharose beads that had been pre-incubated individually with 2 mg of the purified MQIFVKTLTGK-GST, QIFVKTLTGK-GST, and EIFVKTLTGK-GST. The flow-through of antisera was further “positively” selected by overnight incubation with REIFVKTLTGK-GST-conjugated Sulfo-Link Coupling Resin (Cat. No. 20401; Thermo Fisher Scientific, USA) at 4°C. Bound antibodies were eluted by adding 9 ml of 0.1 M glycine HCl (pH 3.5). The eluted antibodies were further incubated with Affi-Gel 10/15 (Cat. No. 1536098; Bio-Rad, USA) that were pre-conjugated with 30 mg of proteins in the extracts of CHY2014 (ate1∆) cells. The unbound anti-^REUb was concentrated by Protein A-Sepharose affinity chromatography (Amicogen, Korea).

Immunoprecipitation of ^REUb and its conjugated proteins

The indicated yeast cells were cultured in 100 ml of YPD medium to A₆₀₀ of ~1. Cells were harvested and washed twice with ice-cold phosphate-buffered saline (PBS). Cell pellets were resuspended in 2 ml of lysis buffer (25 mM Tris-HCl pH 7.4, 150 mM NaCl, 10 mM N-ethylmaleimide [NEM, Cat. No. E3876; Sigma-Aldrich], 10% glycerol, 1% nonidet P40 substitute [NP-40, Cat. No. 11754599001; Sigma-Aldrich], and 1× protease inhibitor cocktail [Cat. No. P8215, Sigma-Aldrich]) and disrupted by bead beating using Mini Beadbeater-24 (BioSpec Products, USA), 10 times for 10 s each at 1 min intervals on ice. Cell extracts were precipitated by centrifugation at 13,000 × g for 15 min at 4°C. The supernatants were incubated with 1 µg of anti-^REUb for 6 h at 4°C and then with 5 µl of Dynabeads Protein A (Cat. No. 10001D; Thermo Fisher Scientific) for 1 h. After washing the beads three times in the lysis buffer, the bound proteins were eluted by boiling in 1× SDS sample buffer and fractionated using 4%-20% Mini-PROTEAN TGX precast protein gels (Cat. No. 4561094; Bio-Rad). Fractionated proteins were transferred into PVDF membranes (polyvinylidene difluoride, Cat. No. 88518; Thermo Fisher Scientific) and ^REUb-conjugated proteins were identified by anti-Ub (P4D1) (Cat. No. Sc-8017; Santa Cruz Biotechnology, USA).

Preparation of SILAC (stable isotope labeling by amino acids in cell culture) samples

S. cerevisiae CHY3188 (arg4∆ lys2Δ) and CHY6056 (ate1Δ arg4Δ lys2Δ) were grown at 30°C to A₆₀₀ of ~1 in 300 ml of SC medium with 20 _µg/ml of either light isotope-labeled ¹²C₆¹⁴N₄-L-arginine (Cat. No. ULM-8347; Cambridge Isotope Laboratories, USA) and ¹²C₆¹⁴N₂-L-lysine (Cat. No. ULM-8766; Cambridge Isotope Laboratories) or 20 _µg/ml heavy isotope-labeled ¹³C₆¹⁵N₄-L-arginine (Cat. No. CNLM-539-H; Cambridge Isotope Laboratories) and ¹³C₆¹⁵N₂-L-lysine (Cat. No. CNLM-291-H; Cambridge Isotope Laboratories). The yeast strains were treated with 2 mM H₂O₂ for 1 h, followed by mixing an equal amount of each culture and subsequent centrifugation at 2,000 × g and 4°C for 10 min. Cell pellets were washed twice with ice-cold PBS, resuspended in 5 ml of lysis buffer (25 mM Tris-HCl pH 7.4, 150 mM NaCl, 10 mM NEM, 10% glycerol, 1% NP-40, 1× protease inhibitor cocktail), and disrupted by bead beating using Mini Beadbeadbeater-24, 10 times for 10 s each at 1 min intervals on ice. Cell lysates were centrifuged at 13,000 × g and 4°C for 15 min. The resulting supernatants were incubated with 2 _µg of anti-^REUb at 4°C for 6 h before adding 10 µl of Dynabeads-Protein A for an additional 1 h of incubation. After washing three times with 1 ml of the yeast lysis buffer, the binding proteins were eluted by boiling in 30 µl of 1× SDS-polyacrylamide gel electrophoresis (PAGE)-sample buffer. The eluted proteins were fractionated by 4%-20% Mini-PROTEAN TGX precast protein gels (Cat. No. 4561096; Bio-Rad) and stained with Coomassie Brilliant Blue (CBB).

Liquid chromatography and tandem mass spectrometry (LC-MS/MS)

The CBB-stained protein bands of interest were excised, destained in 50 mM ammonium bicarbonate with 50% acetonitrile (ACN), and dehydrated with 100% ACN, followed by drying in a vacuum evaporator. The destained proteins were proteolyzed with trypsin by the in-gel digestion method (Kim et al., 2013). The tryptic peptides were reconstituted in 7 µl of 0.1% formic acid, and a 5 µl aliquot was injected into a reverse-phase EASY-Spray PepMap RSLC C18 LC column (0.075 mm inner diameter × 500 mm length) on the Eksigent NanoLC Ultra system with an integrated column heater at 40°C; the column was pre-equilibrated with 96% buffer A (0.1% formic acid in water) and 4% buffer B (0.1% formic acid in ACN). The peptides were eluted with a 4%-35% gradient of buffer B over 150 min and a 32%-80% gradient of buffer B over 40 min. The total ion acquisition run time was set to 190 min at a flow rate of 250 nl/min. A Q-Exactive mass spectrometer (Thermo Fisher Scientific) was operated in data-dependent acquisition mode for the entire analysis. Full scans (m/z 300-1600) were acquired at a resolution of 70,000 using an automatic gain control (AGC) target value of 1e6 and a maximum ion injection time of 30 ms. Tandem mass spectra were generated for up to 12 precursors by high-energy collision dissociation using a normalized collision energy of 35%. The dynamic exclusion was set to 60 s. Fragment ions were detected in normal scan mode using an AGC target value of 5e4 and a maximum ion injection time of 120 ms. Source ionization parameters were as follows: spray voltage, 1.9 kV; capillary temperature, 275°C; and S-Lens RF Level, 50.

Proteomic analysis of SILAC-based LC-MS/MS data

The RAW files from Q-Exactive were directly conveyed to the proteomic data analysis program suite, Proteome Discoverer v2.2 (Thermo Fisher Scientific). We employed the adjusted exemplar workflow for the SILAC experiment and the basic Sequest HT search module. Parameters for the Sequest search module were as follows: enzyme, specifically trypsin; protein database, UniProt reference proteome database of yeast (UP000002311, released in 04/2019) plus modified cRAP contaminant database with accessions in the dust/contact category ( http://www.thegpm.org/crap/); fragment mass tolerance, 0.05 Da; precursor mass tolerance, 20 ppm; dynamic modifications, acetylation of protein-N-term (+42.010565 Da), oxidation of methionine (+15.9949 Da), label: ¹³C₆¹⁵N₂ at lysine (+8.014199 Da), label: ¹³C₆¹⁵N₄ at arginine (+10.008269 Da), Gly–Gly at Lys (Kε-GG) or protein-N-term (+114.042927 Da); and static modification, carbamidomethylation of cysteine (+57.021464 Da). Next, the identified spectra were validated using Percolator module, which was set to use a concatenated target/decoy strategy with a target cut-off q-value of 0.01 at PSM (peptide spectrum match) level. To integrate chromatographic features of identified SILAC pairs, we used Minora Feature Detector module with maximal trace retention time window of 5 min. Peptide abundance from the feature detector module was normalized with total sum of the abundance values over all identified peptides. Each result of the replicates was treated separately.

Identification of mass spectrum matched to the Nt-peptide of ^REUb

The generated mass spectrum files were directly matched to the peptide sequence in the customized ^REUb protein fasta file using the MS-GF+ search algorithm (v20190418) (Kim and Pevzner, 2014). The ^REUb sequence is a virtual one that replaces the two N-terminal residues of Ub MQ with RE. The search engine settings were as follows: trypsin enzyme specificity with 1 as the number of tolerable termini; 20 ppm for MS1 tolerance; variable modifications: oxidation of methionine (+15.9949 Da), label: ¹³C₆¹⁵N₂ at lysine (+8.014199 Da), label: ¹³C₆¹⁵N₄ at arginine (+10.008269 Da); and fixed modification: carbamidomethylation of cysteine (+57.021464 Da). All spectra matched to the Nt-peptide of the ^REUb were manually inspected. For annotating peptide spectra, we used a freely accessible version of the Interactive Peptide Spectrum Annotator (IPSA) (Brademan et al., 2019).

^MQUb undergoes NME, Nt-deamination, and Nt-arginylation to produce ^REUb

We next attempted to determine the levels of endogenous ^REUb and ^REUb-linked protein species by immunoblotting with the purified anti-^REUb. Notably, however, ^REUb and ^REUb–protein conjugates in wild-type S. cerevisiae were barely detectable under normal growth conditions (data not shown), suggesting that their endogenous levels are most likely near or below the detection limit of the antibody.

Since ^MQUb is a putative substrate of NatB Nt-acetylase that targets MQ, MN, ME, and MD at N-termini (Nguyen et al., 2018; Ree et al., 2018), we conjectured that Nt-acetylation might inhibit the NME-triggered Nt-modifications of ^MQUb. To rule out this possibility, we performed anti-^REUb-based immunoblotting in the extracts from naa20Δ cells that lacked a catalytic subunit of the NatB Nt-acetylase (Ree et al., 2018). Indeed, the high-molecular-weight ^REUb–protein adducts were detected in overnight-grown (stationary-phase) naa20Δ yeast cells (Fig. 2F, lane 4). As expected, the ^REUb and ^REUb–protein adducts were abolished in nta1Δ naa20Δ cells (lacking Nta1 Nt-amidase) and ate1Δ naa20Δ cells (lacking Ate1 arginyltransferase) (Nguyen et al., 2018), whereas total Ub and Ub-conjugated proteins were still detected (Fig. 2F). In sum, we conclude that, in vivo, ^MQUb can be converted into ^REUb through the consecutive reactions of NME, Nt-deamination, and subsequent Nt-arginylation (Fig. 2A).

Upregulation of ^REUb–protein adducts in the stationary-growth phase or by oxidative stress

To detect the ^REUb–protein adducts more effectively and sensitively, we performed anti-^REUb-based immunoprecipitation, followed by immunoblotting with a highly Ub-specific monoclonal anti-Ub (P4D1) antibody, in extracts from naa20Δ cells. The hypersensitive coimmunoprecipitation–immunoblotting assays revealed that those ^REUb–protein adducts were markedly induced in wild-type cells in the stationary-growth phase, but hardly at all in the exponential-growth phase (Fig. 2G; cf, lane 6 vs lane 5). In contrast, the ^REUb–protein adducts were not detected in ate1Δ cells regardless of their growth phase (Fig. 2G; lanes 7 and 8).

Ubr1 is as a key E3 Ub ligase of the Arg/N-degron pathway and directly detects the Nt-Arg of Nt-arginylated proteins via its UBR box (Choi et al., 2010). Interestingly, the levels of ^REUb and ^REUb-protein adducts were discernably decreased in ubr1Δ naa20Δ cells, compared with those in naa20Δ cells, indicating the involvement of Ubr1 in processing ^REUb (Fig. 2H; cf, lane 5 vs lane 6).

We also observed that ^REUb–protein adducts were strongly increased in wild-type cells in the presence of 2 mM H₂O₂, an oxidative stressor (Fig. 2I; lane 7). In agreement with results with the stationary-phase yeast, however, the ^REUb-conjugated proteins were hardly detected in ate1Δ cells, regardless of H₂O₂ treatment (Fig. 2I; lanes 6 and 8). Taken together, these results suggest that the ^REUb and ^REUb–protein adducts are upregulated in yeast cells in the stationary-growth phase or by oxidative stress in an Ate1-dependent manner.

Identification of ^REUb and ^REUb-interacting proteins

Given that oxidative stress increases the levels of ^REUb–protein conjugates, we next sought to identify the ^REUb-attached proteins by employing SILAC in the H₂O₂-treated wild-type cells and ate1Δ cells, followed by anti-^REUb-based coimmunoprecipitation and LC-MS/MS (Fig. 3A).

Figure 3

(A) Scheme for quantitative proteomic analysis employing SILAC, enrichment for REUb, REUb-protein conjugates, and their interacting proteins by coimmunoprecipations with anti-REUb and LC-MS/MS. Wild-type S. cerevisiae (with arg4Δ lys2Δ genetic ...

Out of 541 distinct proteins detected by LC-MS/MS, 83 proteins were identified as direct or indirect ^REUb-binding candidates in the H₂O₂-treated wild-type cells (Fig. 3B). STRING enrichment analysis (for protein–protein interaction networks) indicated the functional associations of these putative ^REUb-interacting proteins with a set of cellular processes including gene transcription, vesicle trafficking, DNA replication, RNA processing, and mitochondrial regulation (Fig. 3C).

The tryptic digestion of Ub-conjugated proteins produces a K-ε-GG (K_GG) Ub isopeptide, which allows mapping of the ubiquitylation sites of substrates by LC-MS/MS (Kirkpatrick et al., 2005). Using this Ub-remnant K_GG signature-based approach, in the H₂O₂-treated yeast cells we identified the (putative) ^REUb-conjugated proteins and determined their ubiquitylation sites: Ub (⁴³LIFAGK_GGQLEDGR⁵⁴); Fdc1 (a ferulic acid decarboxylase, ¹²⁶THILSEEK¹³³_GG); Ino80 (a chromatin remodeling ATPase, ⁹³⁹NVQSELGDK_GGIEIDVLCDLTQR⁹⁵⁹); Rpb3 (an RNA polymerase subunit, ¹⁵⁵LTCVAK_GGK¹⁶¹); Uso1 (a vesicle transport protein, ⁹¹²ITEIK_GGAINENLEEMK⁹²⁶); Sla1 (a cytoskeleton-binding protein, ³⁵⁸GIVQYDFMAESQDELTIK³⁷⁵_GG); Ufd4 (a Ub-fusion decay E3 ligase, ⁵⁴⁹AINDQLIK⁵⁵⁶_GG); Rad57 (a DNA repair protein, ⁴⁰⁵SFK_GGASTIIQR⁴¹⁴); Hda1 (a histone deacetylase, ⁵⁶¹SK_GGLNDELR⁵⁶⁸); and Rpl36a (a ribosome 60S subunit, ⁸⁵AK_GGVEEMNNIIAASR⁹⁸) (Fig. 3D). In this setting, however, it should be noted that ^REUb might be attached not only directly to the identified proteins but also indirectly to them by way of the Ub moiety of the already Ub-tagged proteins.

Strikingly, the exhaust manual survey of the tandem MS/MS database revealed the ^REUb-derived REIFVK Nt-peptide, verifying the actual existence of ^REUb in vivo. To the best of our knowledge, ^REUb and its conjugated proteins have not been documented until this study (Fig. 3E).