Proteome-wide detection of S-nitrosylation targets and motifs using bioorthogonal cleavable-linker-based enrichment and switch technique

Abstract

Cysteine modifications emerge as important players in cellular signaling and homeostasis. Here, we present a chemical proteomics strategy for quantitative analysis of reversibly modified Cysteines using bioorthogonal cleavable-linker and switch technique (Cys-BOOST). Compared to iodoTMT for total Cysteine analysis, Cys-BOOST shows a threefold higher sensitivity and considerably higher specificity and precision. Analyzing S-nitrosylation (SNO) in S-nitrosoglutathione (GSNO)-treated and non-treated HeLa extracts Cys-BOOST identifies 8,304 SNO sites on 3,632 proteins covering a wide dynamic range of the proteome. Consensus motifs of SNO sites with differential GSNO reactivity confirm the relevance of both acid-base catalysis and local hydrophobicity for NO targeting to particular Cysteines. Applying Cys-BOOST to SH-SY5Y cells, we identify 2,151 SNO sites under basal conditions and reveal significantly changed SNO levels as response to early nitrosative stress, involving neuro(axono)genesis, glutamatergic synaptic transmission, protein folding/translation, and DNA replication. Our work suggests SNO as a global regulator of protein function akin to phosphorylation and ubiquitination.

Fig. 1

Schematic representation of Cys-BOOST. The workflow is exemplified for SNO analysis of untreated (n = 3) and GSNO-treated (n = 3) HeLa cell extracts. Cys-BOOST incorporates the following steps: (1) irreversible blocking of all free thiols with IAA, (2) switch of specifically reduced Cys with alkyne-derivatized iodoacetamide reagent (IAA-alkyne), (3) reduction and alkylation of the remaining (endogenously) oxidized Cys, (4) protein digestion with trypsin, (5) TMT 10plex™ labeling of Lys residues and N-termini for multiplexed quantitative analysis, (6) conjugation of the Dde-biotin-azide linker to IAA-alkyne labeled peptides via copper(I)-catalyzed alkyne-azide cycloaddition (CuAAC), (7) affinity binding of the Cys peptides to streptavidin beads via the biotin group of the Dde-biotin-azide linker, (8) release of the target peptides with 2% hydrazine by single step mild chemical cleavage of the Dde bond, (9) on-tip pH 10 fractionation of the eluate before (10) LC-MS/MS analysis

Fig. 2

Total Cys analysis by iodoTMT and Cys-BOOST. a Comparison of iodoTMT and Cys-BOOST workflows for total Cys peptides analysis from 100 μg of HeLa lysate per replicate. b and c pie charts showing the share of the target (Cys containing) and background peptides. d Boxplots of scaled log₂ TMT reporter intensities of all quantified Cys containing peptides display a substantially more accurate and precise quantification for Cys-BOOST as compared with iodoTMT (center line: mean; box limits: upper and lower quartiles; whiskers: correspond to the highest or lowest respective value or if the lowest or highest value is an outlier it is 1.5 × interquartile range (IQR)). e Average TMT reporter intensities of Cys-containing PSMs (error bars show the minimum and maximum of n = 3 replicates) indicate a higher recovery by Cys-BOOST. f Number of Cys containing peptides quantified per μg of HeLa lysate. Source data are provided as a Source Data file

Fig. 3

SNO consensus motifs identified by motif-x v1.2. Significance < 1 × 10⁻⁶, fold increase ≥ 1.59. a Motifs for GSNO non-reactive (R ≤ 1.5), b GSNO mild-reactive (1.5 < R < 6; ANOVA p value ≤ 0.05), and c GSNO hyper-reactive (R ≥ 6; ANOVA p value ≤ 0.05) SNO sites. P values derived from n = 3 biological replicates. Source data are provided as a Source Data file

Fig. 4

The STRING networks of proteins with regulated SNO sites after GSNO treatment of HeLa extracts. N = 3 biological replicates. a Proteins with significantly downregulated SNO sites, b the 500 proteins with the strongest upregulated SNO sites. Both networks indicate the role of SNO in splicing, translation, ubiquitin mediated proteolysis, DNA replication, cellular response to stress, vesicle-mediated transport, GTPase-mediated signaling and aminoacyl-tRNA biosynthesis pathways. Source data are provided as a Source Data file

Fig. 5

SNO proteins relevant in heat shock response and hypoxia. a Structures of HYOU1, heat shock proteins (HSP), and PDIA1 with highlighted SNO sites. The color gradient of the highlighted background of Cys symbolizes the R _{[GSNO]:[Control]} ratio. The yellow highlights on the protein structure correspond to the position of the Cys. Protein structures are obtained from SWISS-MODEL. b SNO sites (red) of CITED4 (novel site detected by Cys-BOOST) and HIF1α (detected formerly) both located at C-termini and putatively involved in activation of hypoxia-responsive genes. Cys 184 of CITED4 was detected as free. c All nine Cys of NOSIP were identified, four of which were free Cys (green) and five were nitrosylated (red). Source data are provided as a Source Data file

Fig. 6

SNO analysis of SNAP-treated and non-treated (control) SH-SY5Y cells. a–b Top 20-enriched Gene Ontology (GO) terms (PANTHER Overrepresentation Test, http://www.pantherdb.org/, p values defined by Fisher’s exact test) of the SH-SY5Y SNO proteome: a molecular function and b biological process. c Volcano plot of SNO peptides in SNAP-treated vs non-treated samples. Green dots (n = 39) represent significantly downregulated (≥ 1.5-fold, ANOVA p value ≤ 0.05) and blue dots (n = 65) significantly upregulated (≥ 1.5-fold, ANOVA p value ≤ 0.05) SNO peptides. d High-confidence STRING network of proteins with significantly changed (≥ 1.5-fold, ANOVA p value ≤ 0.05) SNO levels. Proteins with upregulated SNO sites are marked blue, with downregulated green and with both up- and downregulated red. P values derived from n = 4 biological replicates. Source data are provided as a Source Data file