Quantitative Persulfide Site Identification (qPerS-SID) Reveals Protein Targets of H2S Releasing Donors in Mammalian Cells

Abstract

H2S is an important signalling molecule involved in diverse biological processes. It mediates the formation of cysteine persulfides (R-S-SH), which affect the activity of target proteins. Like thiols, persulfides show reactivity towards electrophiles and behave similarly to other cysteine modifications in a biotin switch assay. In this manuscript, we report on qPerS-SID a mass spectrometry-based method allowing the isolation of persulfide containing peptides in the mammalian proteome. With this method, we demonstrated that H2S donors differ in their efficacy to induce persulfides in HEK293 cells. Furthermore, data analysis revealed that persulfide formation affects all subcellular compartments and various cellular processes. Negatively charged amino acids appeared more frequently adjacent to cysteines forming persulfides. We confirmed our proteomic data using pyruvate kinase M2 as a model protein and showed that several cysteine residues are prone to persulfide formation finally leading to its inactivation. Taken together, the site-specific identification of persulfides on a proteome scale can help to identify target proteins involved in H2S signalling and enlightens the biology of H2S and its releasing agents.

Figure 1. Workflow for the identification of persulfide containing peptides using mass spectrometry.

(a) Cells were subjected to trichloroacetic acid (TCA) precipitation and subsequently, the thiols and persulfides were labelled using iodoacetyl-PEG2-Biotin (IAMBio). After digestion of the proteins (Total), single peptides containing either labelled persulfides or cysteines were enriched and separated from non-cysteine peptides (flow-through) using streptavidin agarose beads. After several washing steps, persulfide containing peptides were eluted using tris(2-carboxyethyl)phosphine (TCEP) where thiol containing peptides were not affected. The subsequent accessible cysteines were labelled with iodoacetamide (IAM) (Elution). As control, labelled thiol peptides remaining on the beads were eluted using 10 mM TCEP and 80% acetonitrile (ACN) (Beads). Samples of the total, flow-through, elution and bead fraction were subjected to liquid chromatography and mass spectrometry (LC-MS/MS) and the peptides were identified using the PEAKS 7.0 proteomics software. (b) Number of identified peptides in the different fractions as described in a). Shown are the mean values of two independent experiments. S-IAMBio: Persulfide peptide modified by IAMBio.

Figure 2. Workflow for the detection of quantitative changes of persulfide formation upon treatment with different H₂S donors.

(a) The experiment was performed as described in Fig. 1a, except that the cells were grown in heavy or light SILAC medium (red/blue) allowing the quantification of induced persulfides. In order to induce persulfide formation the cells were treated 30 min with 200 μM of the polysulfide Na₂S₄ and 1 mM of the slow releasing H₂S reagent GYY4137 for 4 h. Peptides were quantified by MaxQuant. (b) Elution profile of GYY4137 and Na₂S₄ stimulated cells. The experiment was performed four times (two times forward, two times reverse) and the mean values and error bars of significant peptides (p ≤ 0.05) appearing in at least 3 of 4 experiments are plotted as log2 value. The median of the heavy to light ratio (H/L) of non-cysteine peptides in the total fraction was used for normalisation. Dashed line: threshold of peptides that were at least 30% induced (1.3×). (c) Venn diagram of peptides and proteins influenced by GYY4137 and Na₂S₄. (d) Diagram showing the number of persulfide sites per protein changed by treatment with GYY4137 and Na₂S₄.

Figure 3. Amino acid enrichment analysis of persulfide containing peptides and proteomic comparison of persulfides, S-nitrosated cysteines and S-sulfenylated cysteines.

(a) pLogo analysis of enriched amino acids up to 15 positions before or after the reactive cysteine. Red: negative charged amino acids, blue: positive charged amino acids, black: hydrophobic amino acids (b) pLogo analysis of enriched amino acids 15 positions before or after cysteines that were not affected by H₂S stimulation found in the beads fraction. (c) Venn diagram comparing proteins that were S-sulfhydrated (R-S-SH, persulfides) by GYY4137 and Na₂S₄ as well as S-sulfenylated (R-S-OH) and S-nitrosated (R-S-NO). (d) Venn diagram comparing S-sulfhydrated (R-S-SH, persulfides), S-sulfenylated (R-S-OH) and S-nitrosated peptides (R-S-NO).

Figure 4. Protein enrichment analysis of persulfides affected by GYY4137 and Na₂S₄.

(a) Cellular compartment enrichment analysis using the DAVID program. (b) GO term enrichment analysis of biological processes (GOTERM_BP), molecular function (GOTERM_MF) and KEGG pathway enrichment analysis.

Figure 5. In vitro characterisation of persulfide formation on PKM2.

(a) Structure showing the human PKM2 monomer and tetramer (PDB: 3SRH). Cysteine peptides forming persulfides are highlighted (yellow spheres; A, B, C, D). Black: intersubunit contact site; letters in square brackets: sequence of misscleaved peptide. (b) Purified PKM2 from rabbit muscle was incubated with increasing concentrations of Na₂S₄, Na₂S, NaSH and GYY4137. PKM2 activity (μmol/min/mg) was measured in a coupled enzyme assay with LDH as second enzyme monitoring the consumption of NADH at 340 nm. (c) The activity of PKM2 was followed as decline in absorption at 340 nm. After 1 min 5 mM pyruvate was added to bypass the reaction catalysed by PKM2. (d) The experiment was carried out as described in (b) except that 1 mM DTT was added in parallel to treatment with 200 μM Na₂S₄. Data are means +/− SD, **p < 0.01 Na₂S₄ vs Na₂S₄ + DTT, ***p < 0.001 Ctrl as well as DTT vs Na₂S₄. (e) Workflow to confirm that persulfides are formed at PKM2. PKM2 was incubated with 100 μM and 500 μM Na₂S₄ or kept untreated. Induced persulfides were modified with iodoTMT similar to qPerS-SID. After digestion with trypsin the persulfide peptides were enriched using an anti-iodoTMT resin and subjected to TCEP elution followed by IAM blocking as described for the proteomic approach. In parallel, direct labelled persulfides (S-iodoTMT) were eluted using iodoTMT elution buffer. The eluted peptides were subjected to LC-MS/MS measurement and the peptides were identified using PEAKS 7.0. (f) Spectra counts of iodoTMT labelled cysteine peptides (iodoTMT), persulfide peptides identified according to the qPerS-SID protocol (TCEP elution, IAM) and iodoTMT labelled persulfide peptides (TMT elution, S-iodoTMT).