Abstract

   Acute inflammatory responses often involve the production of reactive
   oxygen and nitrogen species by innate immune cells, particularly
   macrophages. How activated macrophages protect themselves in the face
   of oxidative-inflammatory stress remains a long-standing question.
   Recent evidence implicates reactive sulfur species (RSS) in
   inflammatory responses; however, how endogenous RSS affect macrophage
   function and response to oxidative and inflammatory insults remains
   poorly understood. In this study, we investigated the endogenous
   pathways of RSS biogenesis and clearance in macrophages, with a
   particular focus on exploring how hydrogen sulfide (H[2]S)-mediated
   S-persulfidation influences macrophage responses to
   oxidative-inflammatory stress. We show that classical activation of
   mouse or human macrophages using lipopolysaccharide and interferon-γ
   (LPS/IFN-γ) triggers substantial production of H[2]S/RSS, leading to
   widespread protein persulfidation. Biochemical and proteomic analyses
   revealed that this surge in cellular S-persulfidation engaged ∼2% of
   total thiols and modified over 800 functionally diverse proteins.
   S-persulfidation was found to be largely dependent on the cystine
   importer xCT and the H[2]S-generating enzyme cystathionine γ-lyase and
   was independent of changes in the global proteome. We further
   investigated the role of the sulfide-oxidizing enzyme sulfide quinone
   oxidoreductase (SQOR), and found that it acts as a negative regulator
   of S-persulfidation. Elevated S-persulfidation following LPS/IFN-γ
   stimulation or SQOR inhibition was associated with increased resistance
   to oxidative stress. Upregulation of persulfides also inhibited the
   activation of the macrophage NLRP3 inflammasome and provided protection
   against inflammatory cell death. Collectively, our findings shed light
   on the metabolism and effects of RSS in macrophages and highlight the
   crucial role of persulfides in enabling macrophages to withstand and
   alleviate oxidative-inflammatory stress.

   Keywords: M1 macrophages, Persulfides, Oxidative stress, Inflammation,
   Cell death

1. Introduction

   Innate inflammatory responses are characterized by the production of
   reactive oxygen and nitrogen species (ROS and RNS) thus being closely
   associated with oxidative/nitrosative stress [[31]1,[32]2]. Innate
   immune cells, in particular macrophages, utilize ROS/RNS, such as
   hydrogen peroxide (H[2]O[2]) and nitric oxide (NO), to kill microbes
   and tumor cells. In addition to their cytotoxic function, ROS/RNS act
   as versatile signaling molecules in macrophages and other inflammatory
   cells, where they regulate diverse processes including metabolism,
   motility, cell division and differentiation [[33][3], [34][4],
   [35][5]]. Recent evidence shows that hydrogen sulfide (H[2]S) and other
   reactive sulfur species (RSS) are also generated during inflammatory
   processes [[36][6], [37][7], [38][8], [39][9]]. The evidence further
   suggests that similar to ROS and RNS, RSS also exert regulatory effects
   on cellular inflammatory responses [[40][6], [41][7], [42][8],
   [43][9]]. However, while the roles and mechanisms of action of innate
   immune cell-derived ROS/RNS have been extensively studied and well
   characterized, those of endogenous RSS have been less studied and
   characterized.

   In recent years, the cellular mechanisms of H[2]S synthesis and
   metabolism have been delineated in great detail [[44][10], [45][11],
   [46][12], [47][13]]. H[2]S is synthesized in cells by cystathionine
   γ-lyase (CSE) and cystathionine β-synthase (CBS) in the
   transsulfuration pathway, and 3-mercaptopyruvate sulfurtransferase
   (MPST), involved in cysteine (Cys) catabolism. H[2]S is primarily
   metabolized via the mitochondrial sulfide oxidation pathway. The first
   enzyme in this pathway is sulfide quinone oxidoreductase (SQOR), which
   catalyzes the oxidation of H[2]S to reactive persulfide species and
   reduces coenzyme Q (CoQ). The biochemical interplay between H[2]S and
   various RSS has become a topic of increased research interest. Among
   RSS, sulfane sulfur (S^0) species, such as hydropersulfides (RSSH) and
   polysulfides (RSS(n)SR, n ≥ 1), are increasingly recognized as
   potential biological mediators [[48][14], [49][15], [50][16]]. In this
   regard, evidence indicates that the above-mentioned H[2]S biosynthetic
   and degrading enzymes influence endogenous persulfide levels; however,
   the precise contribution of each enzyme remains to be fully explored,
   and it may vary depending on the cell type and context. Furthermore,
   despite their emerging importance, there is limited knowledge about the
   metabolism and roles of persulfides in macrophages and related cell
   types.

   Protein S-persulfidation (P-SSH), the process in which protein Cys
   thiols are converted to persulfides, has recently emerged as an
   important mechanism of H[2]S-based signaling [[51]17,[52]18]. The
   available evidence indicates that persulfidation of proteins serves two
   main roles: (1) the reversible modulation of protein function, and (2)
   the protection of protein thiols against overoxidation [[53][17],
   [54][18], [55][19], [56][20]]. Persulfidation has been shown to
   modulate the activity of various classes of proteins, including
   transcription factors, proteases, kinases/phosphatases, and others
   [[57]17,[58][21], [59][22], [60][23], [61][24]]. Nevertheless, the
   mechanisms that govern persulfide formation in cells are incompletely
   understood, and the cell-specific roles of protein persulfidation
   largely remain to be characterized. The kinetics and magnitude of
   S-persulfidation are determined by several factors, including cellular
   H[2]S fluxes and the redox milieu. Yet the exact mechanisms by which
   proteins undergo persulfidation in cells are not well defined and are a
   subject of intensive research [[62][14], [63][15], [64][16], [65][17],
   [66][18],[67]25]. In general, de-novo, non-enzymatic protein
   persulfidation is thought to occur via one of two mechanisms: (i) the
   reaction of oxidized Cys derivatives with sulfide, or (ii) sulfide
   oxidation products reacting with Cys thiols. In addition, recent
   findings support enzyme-catalyzed protein persulfidation
   [[68]26,[69]27]. In this context, the role of SQOR is of particular
   interest. On the one hand, SQOR maintains low levels of H[2]S and can
   thus limit the generation of P-SSH; conversely, SQOR-catalyzed H[2]S
   oxidation in itself is source of persulfides, in particular glutathione
   persulfide (GSSH), which could then promote P-SSH formation [[70]28].
   These divergent scenarios remain to be tested in different cell types
   and conditions. In particular, the role of SQOR in regulating
   persulfidation processes in macrophages is so far unknown, and the
   potential functional consequences of such regulation remain to be
   determined.

   In this study, we investigated the formation, metabolism and roles of
   H[2]S/persulfides in M1/classically-activated macrophages. Our results
   reveal that activation of human and mouse macrophages induces robust
   H[2]S generation and widespread S-persulfidation. We provide evidence
   that SQOR negatively regulates endogenous S-persulfidation and show
   that elevated levels of persulfides are associated with enhanced
   macrophage survival in response to oxidant or inflammatory stress.
   Altogether, our findings indicate that the upregulation of persulfides
   in activated macrophages serves as a self-protective mechanism to
   mitigate and cope with oxidative and inflammatory stress.

2. Materials and methods

2.1. Antibodies and reagents

   The following primary antibodies were used in this study: anti-SQOR
   (17256-1-AP) and anti-SLC7A11/xCT (26864-1-AP) from Proteintech,
   anti-caspase 1 (SC-515) from Santa Cruz Biotechnology, anti-NLRP3
   (ag-20b-0014) and anti-ASC (AG-25B-0006) from AdipoGen, anti-IL-1β
   (5129-100) from Biovision, anti-GAPDH (ab9484) from Abcam. LPS from E.
   coli 055:B5 (L4005), PMA (P8139), dl-propargylglycine (p7888),
   aminooxyacetate ([71]C13408) and sulfasalazine (S0883) were purchased
   from Sigma. Mouse IFN-γ (485-MI-100) and human IFN-γ (285-IF-100) were
   from R&D Systems. Poly (dA:dT) (tlrl-patn-1) and flagellin (tlrl-stfla)
   were from InvivoGen. EZ-Link Maleimide-PEG2-Biotin (21901BID),
   streptavidin-agarose resin (20361) and disuccinimidyl suberate (DSS)
   (21655) were from Thermo Scientific. The fluorescent probes WSP-5 and
   SSP4 were a kind gift of Prof. Ming Xian (Brown University). The MPST
   inhibitor I3-MT-3 was a kind gift of Prof. Kenjiro Hanaoka (Keio
   University, Tokyo, Japan). Tissue culture media and reagents were from
   Biological Industries (Beit Haemek, Israel).

2.2. Preparation of Cys persulfide solution

   A Cys persulfide (Cys-SSH) solution was prepared based on published
   protocols for persulfide synthesis by reacting cystine with Na[2]S
   [[72][29], [73][30], [74][31]]. In brief, freshly prepared degassed
   solutions of 20 mM cystine (pH 8.8) and 40 mM Na[2]S were mixed 1:1 and
   incubated for 30 min at 37 °C and then promptly aliquoted and frozen at
   −20 °C. The solution sulfane sulfur content was determined by cold
   cyanolysis conducted at pH 8.8 and colorimetric detection of the
   resulting ferric thiocyanate complex [[75]32]. Under these conditions,
   cystine alone did not generate any detectable sulfur product in the
   cold cyanolysis assay. According to literature [[76]33], the Cys-SSH
   solution thus prepared contains several species in metastable
   equilibrium, namely, Cys-SSH, Cys–SSS–Cys, cystine, Cys, and H[2]S.
   Absorbance measurements at 335 nm [[77]31,[78]32] indicated that
   Cys-SSH was the major species in the solution. Furthermore, when the
   solution was kept on ice, no appreciable absorbance changes were seen
   for at least 1 h. Regardless, the Cys-SSH solution was used immediately
   after the preparation or immediately after thawing from the frozen
   state.

2.3. Cell culture and treatments

   In this work, we have used human and mouse monocytes/macrophages as
   follows. THP-1 cells (human monocyte cell line) were obtained from
   American Type Culture Collection (ATCC) and maintained in RPMI 1640,
   supplemented with 10% (v/v) fetal bovine serum (FBS), 1%
   penicillin/streptomycin and 1% sodium pyruvate. The cells were
   sub-cultured using fresh media every 2–4 days and maintained in an
   incubator (37 °C, 5% CO[2]). THP-1 monocytes were differentiated to
   adherent macrophages (TDM) by incubation with PMA (166 nM) for 48 h at
   37 °C. THP-1 cell differentiation was enhanced by removing the
   PMA-containing media and adding fresh media for a further 24 h. The
   mouse macrophage cell line J774 was obtained from ATCC and maintained
   in DMEM containing 10% FBS, 1% penicillin/streptomycin, 1% sodium
   pyruvate and 1% glutamine, at 37 °C in 5% CO[2].

   Bone marrow-derived macrophages (BMDM) from 8- to 12-week-old C57BL/6
   male mice were generated using established protocols. In brief, femurs
   of mice were flushed, and a single-cell suspension of the bone marrow
   was prepared in DMEM medium. The suspension was centrifuged at 1500 rpm
   for 5 min to pellet the cells. The cells were then resuspended in DMEM
   supplemented with 30% (v/v) conditioned L929 (ATCC number: CCL-1)
   medium at a density of 8x10^6 cells/well. On the seventh day of
   culture, the cells were transferred to a new plate and resuspended in
   DMEM supplemented with 10% FBS, 1% penicillin/streptomycin and 1%
   glutamine, at 37 °C in 5% CO[2].

   For the induction of M1 macrophages, cells were stimulated for 24 h
   with LPS (0.5 μg/ml) and IFN-γ (50 ng/ml or 20 ng/ml for mouse and
   human cells, respectively). In order to activate the NLRP3
   inflammasome, macrophages and monocytes were exposed to LPS (0.5 μg/ml)
   for 4 h, followed by stimulation with 10 μM nigericin or 5 mM ATP. In
   experiments assessing the activation of caspase-1 and IL-1β, cell
   stimulation was performed in serum-free-medium. In order to activate
   the NLRC4 and AIM2 inflammasomes LPS-primed macrophages were
   transfected with flagellin (1 μg/ml) or poly (dA:dT) (2 μg/ml),
   respectively, using Lipofectamine RNAiMAX (Thermo Fisher Scientific).
   Procaspase-1 and pro-IL-1β in the cell extracts and their active forms
   secreted in the culture supernatant precipitated with 10%
   trichloroacetic acid were analyzed by immunoblotting.

2.4. Measurement of intracellular cysteine

   The intracellular Cys content was determined by Gaitonde's acid
   ninhydrin method [[79]34]. In brief, after treatments, the cells were
   washed with cold PBS, and then 5% sulfosalicylic acid (SSA) was added
   to each well. The cells were collected and centrifuged for 15 min at
   15000 g. The pellets were used for protein quantification and the
   supernatants were used for the assay. The pH of SSA supernatant was
   adjusted to 8.3 using NaOH and then the samples were reduced with 5 mM
   DTT for 15 min. Next, 0.1 ml of glacial acetic acid was added to 100 μl
   sample followed by addition of 100 μl of acid ninhydrin reagent. After
   heating (100 °C for 10 min) and cooling, the samples were diluted 1:2
   with 100% ethanol. Sample absorbance was measured at 560 nm, and Cys
   levels were quantified using Cys standards dissolved in 5% SSA and
   processed in the same manner as the samples.

2.5. Measurement of intracellular H[2]S, sulfane sulfur and ROS

   Intracellular H[2]S and S^0 levels were evaluated using the fluorescent
   dyes WSP-5 and SSP4, respectively. After treatments, the cells were
   loaded with the probes (50 μM WSP-5 or 20 μM SSP4) in the presence of
   500 μM hexadecyltrimethylammonium bromide (CTAB) for 30 min (for WSP-5)
   or 15 min (for SSP4), at 37 °C, in the dark. Thereafter, the cells were
   washed with PBS and transferred to FACS tubes for analysis. The
   fluorescence intensity was acquired by LSRFortessa™ (BD Biosciences)
   and flow cytometry data were analyzed using FCS express software. For
   fluorescence quantification, samples were acquired in triplicate, and
   10,000 events were registered from each sample. Intensity of specific
   fluorescence was expressed as the median fluorescence intensity (MFI),
   from the average of at least three repeated experiments.

   To provide a general assessment of ROS, the cell permeable fluorescent
   dye, 2′,7′-dichlorofluorescin diacetate (H2DCFDA) was employed as
   detailed before [[80]35]. In brief, the cells were loaded with
   10 μM H2DCFDA and incubated for 60 min at 37 °C. After a wash step, the
   cells were exposed to treatments in serum-free and phenol red-free
   medium and fluorescence was detected using a microplate reader at
   493 nm excitation and 523 nm emission.

2.6. H[2]S consumption assay

   After treatments, cells were washed twice with PBS and then collected
   in 1 ml of buffer A (PBS/20 mM HEPES, 5 mM d-glucose, pH 7.5) into
   pre-weighed microfuge tubes. Then, the cells were centrifuged (1,700 g,
   5 min, 4 °C), the supernatants were removed and cell pellets were
   resuspended in 20 vol of buffer A. The cell suspensions were then
   incubated with fresh Na[2]S solution (100 μM) prepared in 0.2 M HEPES
   (pH 7.5) and incubated at 37 °C for 15 min. Thereafter, the cells were
   centrifuged (3,000 g, 1 min, 4 °C) and 50 μl of the supernatant was
   mixed with 50 μl of 25 μM 7-Azido-4-Methylcoumarin (AzMC) fluorescent
   dye (Sigma, 802409) in a 96-well microplate and incubated for 30 min at
   room temperature. Finally, the samples were analyzed using fluorescence
   plate reader (Ex: 356 nm, Em: 450 nm). The sulfide concentrations were
   determined by comparison of the fluorescence intensity to a standard
   curve prepared using Na[2]S (0–100 μM).

2.7. Assessment of protein persulfidation using the biotin thiol assay

   The BTA was performed according to the original protocol of Gao et al.
   [[81]36] with minor modifications. In brief, whole cell lysates were
   prepared using RIPA buffer (100 mm Tris-HCl, 150 mm NaCl, 1 mm EDTA,
   0.5% Triton X-100, 0.5% deoxycholic acid, with protease inhibitors, pH
   7.5) and protein concentrations were adjusted to 0.8 mg/ml. A total of
   1 mg protein was used for each reaction. Thiol and persulfide groups
   were modified with EZ-Link Maleimide-PEG2-Biotin at a final
   concentration of 0.1 μM for 30 min at room temperature. After acetone
   precipitation, the protein pellets were washed with 70% cold acetone
   three times, and then resuspended in resuspension buffer (50 mM
   Tris–HCl, 150 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, 1% SDS, pH 7.5).
   Thereafter, biotinylated proteins were captured by streptavidin-agarose
   resin overnight at 4 °C. After extensive washes of the beads,
   disulfide-linked proteins (formerly P-SSH) were eluted with 20 mm DTT
   for 30 min at room temperature as described in detail in Ref. [[82]36].
   Eluates were subjected to SDS-PAGE or immunoblotting as indicated in
   the figures.

2.8. Sample preparation for proteomic analysis

   Persulfide and total proteome analysis was performed using THP-1 cells
   to allow for the scalability needed for a proteome-wide analysis.
   Persulfidated proteins were isolated using the BTA method detailed
   above. The proteins in the gel were reduced with 3 mM DTT (60 °C for
   30 min), modified with 10 mM iodoacetamide in 100 mM ammonium
   bicarbonate (in the dark, room temperature for 30 min) and digested in
   10% acetonitrile and 10 mM ammonium bicarbonate with modified trypsin
   (Promega) at a 1:10 enzyme-to-substrate ratio, overnight at 37 °C. An
   additional second trypsinization was done for 4 h at a 1:20
   enzyme-to-substrate ratio.

2.9. Liquid chromatography–MS/MS

   The tryptic peptides obtained above were desalted using C18 tips
   (homemade stage tips) and were subjected to MS analysis, as follows.
   The peptides were resolved by LC-MS/MS using a Q Exactive HF mass
   spectrometer (Thermo) fitted with a capillary HPLC (Evosep) with EV1137
   column, ID: 150 μm, ReproSil-Pur C18, 1.5 μm beads (Dr Maisch, GmbH,
   Germany). The peptides were separated with the built-in 15 samples per
   day method (88 min gradient). MS was performed in a positive mode using
   repetitively full MS scan (m/z 300–1,500, resolution 120,000 for MS1
   and 15,000 for MS2) followed by 20 data-dependent MS/MS scans at 27%
   normalized collision energy (HCD). The automatic gain control settings
   were 3x10^6 for survey MS scans and 1x10^5 for MS/MS scans. A dynamic
   exclusion list was enabled with exclusion duration of 20 s.

   Tryptic peptides from input samples were resolved by reverse-phase
   chromatography on 0.075 X 180-mm fused silica capillaries (J&W) packed
   with Reprosil reversed phase material (Dr Maisch GmbH, Germany). The
   peptides were eluted with linear 180-min gradient of 5–28% acetonitrile
   with 0.1% of formic acid followed by 15-min gradient of 28–95% and
   25 min at 95% acetonitrile with 0.1% formic acid in water at flow rates
   of 0.15 μl/min. MS/MS analysis was performed using the Q Exactive HFX
   mass spectrometer (Thermo) in a positive mode using the same settings
   as above, expect for selecting the 30 most dominant ions from the first
   MS scan.

2.10. Proteomic data analysis

   The mass spectrometry data were analyzed using the MaxQuant software
   (version 1.5.2.8) [[83]37,[84]38] for peak picking and identification
   using the Andromeda search engine, searching against the human UniProt
   database with mass tolerance of 6 ppm for the precursor masses and
   20 ppm for the fragment ions. Oxidation on methionine and protein
   N-terminus acetylation were accepted as variable modifications and
   carbamidomethyl on Cys was accepted as static modifications. Minimal
   peptide length was set to 7 amino acids and a maximum of two
   miscleavages was allowed. Peptide- and protein-level false discovery
   rates (FDRs) were filtered to 1% using the target-decoy strategy.
   Protein tables were filtered to eliminate the identifications from the
   reverse database, and common contaminants. Statistical analysis was
   performed using the Perseus software platform (version 1.6.7.0)
   [[85]39]. Proteins showing at least 4-fold change of intensities
   between conditions with p-value below 0.05 were considered as
   differentially persulfidated. A smaller list of differentially
   persulfidated proteins (p < 0.01) were subjected to bioinformatic
   analysis. Functional enrichment analysis was performed using Matecaspe
   ([86]http://metascape.org) with the Custom Analysis module (Minimum
   Overlap: 3, p-value Cutoff: 0.01, Minimum enrichment: 1.5) [[87]40].
   The enrichment background list included a total of 3717 genes, obtained
   from the total proteome analysis.

2.11. Assessment of cellular persulfide content using a lead acetate-based
assay

   Cells were extracted with NP-40 containing hypotonic buffer (10 mM
   HEPES, 10 mM KCl, 0.1 mM EDTA, 3 mM MgCl[2], 0.5 % NP-40, with protease
   inhibitors, pH 7.5), centrifuged to remove cellular debris, and diluted
   to 2 mg/ml using HEN buffer (25 mM HEPES, 1 mM EDTA, 10 μM neocuproine,
   pH 7.5). Next, equal amount of samples (∼3 mg protein in 600 μL) were
   incubated with 0.1% MMTS for 15 min at room temperature with rotation.
   Thereafter, protein samples and Cys-SSH standard solutions were placed
   in a round bottom deep-well plate and DTT (1 mM) was added to each
   well, followed by a 2-h incubation at 37 °C with gentle shaking. The
   released sulfide was subsequently detected and quantified using a lead
   acetate embedded filter paper, essentially as described by Hine and
   Mitchell [[88]41]. In some experiments, low- and high-mass persulfides
   were isolated by centrifugal filtration of lysates using an Amicon
   Ultra-4 10 kDa centrifugal concentrator (Millipore), where the low-mass
   fraction represented the flow through and high-mass fraction was the
   retentate. The sample fractionation was performed essentially as
   described elsewhere [[89]42].

2.12. siRNA knockdown

   Knockdown of SQOR was accomplished by electroporation of small
   interfering RNA (siRNA) oligos. The siRNA oligos were purchased from
   Horizon Discovery and included the following: ON-TARGETplus
   Non-targeting Control Pool (D-001210-02) and ON-TARGETplus Human SQOR
   siRNA (J-008271-10 and J-008271-12). The siRNA oligos (3 μM) were
   electroporated into 4x10^6 THP-1 cells in 100 μL Opti-MEM (275 V,
   1.5 ms, NEPA21, Nepa Gene Co., Ltd., Japan). Analyses were performed
   72 h after electroporation. In some experiments, a second
   electroporation was conducted after 24 h, and cells were then cultured
   for 48 h.

2.13. Assessment of cell death

   Cells were plated at 40,000 cells/well in 96-well plates and grown
   overnight. After a medium change, the cells were treated as indicated
   in the figures legends. Cell death was evaluated using lactate
   dehydrogenase (LDH) release assay, which is based on the principle that
   upon permeabilization of plasma membrane, LDH, which is a cytosolic
   enzyme, is released into the culture supernatants. LDH release into the
   medium was measured using the Cytotox96 cytotoxicity kit as per
   manufacturer instructions (Promega).

2.14. Assessment of ASC oligomerization and speck formation

   ASC oligomerization was analyzed using a method adapted from Shi et al.
   [[90]43]. After treatment with indicated stimuli, cells were washed
   with cold PBS, resuspended in a PBS buffer containing 1% NP-40 and
   EDTA-free protease inhibitor cocktail, and lysed by shearing 10 times
   through a 21G needle. Nuclei and unlysed cells were removed by
   centrifugation at 250g for 5 min. The cell lysates were then
   centrifuged at 6000 g for 15 min at 4 °C, and the pellets and
   supernatants were used as the NP-40-insoluble fractions and
   NP-40-soluble fractions, respectively. For the detection of ASC
   oligomerization, the NP-40-insoluble pellets were washed twice with PBS
   and then were resuspended in 300 μl of the lysis buffer. The
   resuspended pellets underwent crosslinkage for 30 min at room
   temperature with 2 mM disuccinimidyl suberate (DSS) and then were
   centrifuged at 3500 g for 5 min at 4 °C. The supernatants were
   discarded and the pellets were dissolved in 2X SDS sample buffer.
   Finally, the samples were boiled at 95 °C for 10 min, centrifuged at
   16,000 g for 2 min, and the supernatants were used for immunoblotting
   with anti-ASC antibody. For ASC speck formation, macrophages were
   seeded on chamber slides and allowed to attach overnight. The next day,
   the cells were treated with indicated stimulators. The cells were then
   fixed in 4% paraformaldehyde followed by permeabilization with 0.3%
   Triton X-100 for 10 min. The slides were blocked with 5% bovine serum
   albumin, followed by ASC and DAPI staining. Images were acquired on a
   Confocal LSM 880 microscope.

2.15. Statistics

   All data are presented as the mean ± S.D. Statistical differences were
   analyzed by one-way analysis of variance (ANOVA) using Prism software
   (GraphPad). A p-value <0.05 was considered statistically significant.

2.16. Data availability

   The mass spectrometry proteomics data have been deposited to the PRIDE
   Archive ([91]http://www.ebi.ac.uk/pride/archive/) via the PRIDE partner
   repository with the data set identifier PXD044043.

3. Results

3.1. Enhanced H[2]S/RSS production in M1-activated monocytes/macrophages

   We initially set out to examine the biosynthesis of H[2]S/RSS in
   response to stimulation of macrophages with lipopolysaccharide and
   interferon-γ (LPS/IFN-γ), which drives classical activation of
   macrophages and polarizes them towards an M1-like phenotype [[92]44].
   We first measured levels of Cys, a major precursor of H[2]S/RSS, under
   basal conditions and in response to LPS/IFN-γ stimulation. We found
   that Cys levels were significantly increased (∼2 fold) in LPS/IFN-γ
   stimulated THP-1 cells (human monocytes) or J774 cells (mouse
   macrophages) compared to non-stimulated cells ([93]Fig. 1A). This
   effect was abolished by cotreatment with sulfasalazine (SSZ), an
   inhibitor of the cystine transporter, xCT ([94]Fig. 1A). We also
   observed that xCT expression was increased in response to LPS/IFN-γ
   stimulation of these cells ([95]Fig. 1B). Next, we used the fluorescent
   probes WSP-5 and SSP4 to assess cellular levels of H[2]S and S^0
   species, respectively [[96]45,[97]46]. We found that LPS/INF-γ
   stimulation triggered a significant increase in H[2]S and S^0 levels in
   THP-1 cells, THP-1-derived macrophages (TDM), J774 macrophages and
   primary bone marrow derived mouse macrophages (BMDM) ([98]Fig. 1C and D
   and [99]Supplementary Fig. S1).

Fig. 1.

   [100]Fig. 1
   [101]Open in a new tab

   Increased generation of reactive sulfur species in LPS/IFN-γ-activated
   monocytes/macrophages. (A) THP-1 monocytes or J774 macrophages were
   either left untreated or treated for 24 h with LPS/IFN-γ in the absence
   or presence of sulfasalazine (SSZ), followed by determination of
   intracellular Cys content. (B) THP-1 or J774 cells were treated for
   24 h with LPS/IFN-γ followed by analysis of xCT expression by
   immunoblotting. (C) THP-1 monocytes, THP-1-derived macrophages (TDM),
   J774 cells, or bone-marrow-derived macrophages (BMDM) were either left
   untreated or treated for 24 h with LPS/IFN-γ and cellular H[2]S levels
   were assessed by fluorescent dye WSP-5 staining followed by flow
   cytometry analysis. (D) Cells were treated as in C and S^0 levels were
   assessed by fluorescent dye SSP4 staining followed by flow cytometry
   analysis. Graphs (A, C, D) show mean ± SD (n ≥ 3). MFI, mean
   fluorescence intensity.

3.2. Widespread protein S-persulfidation in M1-activated macrophages

   We next focused our attention to H[2]S-mediated protein modification,
   i.e., persulfidation. Using the biotin thiol assay (BTA) of
   persulfidation analysis [[102]36] we observed a substantial increase in
   protein persulfidation in M1-activated THP-1 cells and TDM ([103]Fig.
   2A). BTA reactions without DTT were performed in parallel as a negative
   control for assay specificity and non-specific binding ([104]Fig. 2A).
   Using TDM and J774 macrophages, we next examined the effects of several
   inhibitors of enzymes/proteins that may affect cellular persulfide
   formation. We found that macrophage treatment with the CSE inhibitor
   dl-propargylglycine (PAG) markedly lowered protein persulfidation,
   while in comparison, the CBS inhibitor aminooxyacetate (AOAA) and the
   MPST inhibitor (I3-MT-3) exerted only minor effects ([105]Fig. 2B).
   Notably, the xCT inhibitor SSZ greatly decreased protein persulfidation
   in the activated macrophages ([106]Fig. 2B). Taken together, the
   results thus far indicate that LPS/IFN-γ activation of macrophages
   leads to increased cystine import and intracellular Cys pools, followed
   by CSE-dependent persulfidation of a large number of proteins.

Fig. 2.

   [107]Fig. 2
   [108]Open in a new tab

   Enhanced S-persulfidation in LPS/IFN-γ-activated macrophages. (A) THP-1
   monocytes or THP-1-derived macrophages (TDM) cells were either left
   untreated or treated for 24 h with LPS/IFN-γ. Thereafter, protein
   persulfidation was analyzed using the biotin thiol assay (BTA). (B) TDM
   or J774 cells were either left untreated or treated for 24 h with
   LPS/IFN-γ ± inhibitors and thereafter protein persulfidation was
   analyzed using BTA. The inhibitors used were dl-propargylglycine (PAG,
   3 mM), I3-MT-3 (10 μM), aminooxyacetate (AOAA, 1 mM) or sulfasalazine
   (SSZ, 400 μM). (C) Schematic illustration of the assay for measuring
   total cellular persulfide content. In this assay, thiols and
   persulfides are first capped with S-methyl methanethiosulfonate (MMTS).
   The samples are then transferred to microwells and incubated with DTT.
   Sulfide released from the nascent Cys-S-S-Me groups is detected using
   lead-acetate paper. (D) TDM were either left untreated or treated with
   LPS/IFN-γ (24 h) or with 50 μM Cys-SSH (15 min). Persulfide content in
   the cell lysates was determined using the lead acetate-coupled assay
   depicted in C. (E) Lysates were prepared and analyzed as in D. Size
   separation was performed using a centrifugal filter with a molecular
   mass cutoff of 10,000 Da.

   We sought to obtain a more quantitative assessment of cellular
   persulfide content in resting/activated macrophages. To this end, we
   developed a simple biochemical assay (schematically depicted in
   [109]Fig. 2C) in which thiols and persulfides are initially blocked
   with S-methyl methanethiosulfonate (MMTS), followed by DTT treatment
   and capture of the released sulfide by lead acetate paper (see Material
   and methods for further details). Using this assay in TDM, we observed
   that persulfide levels were essentially undetectable in lysates
   obtained from untreated cells; in contrast, they were readily detected
   in samples from LPS/IFN-γ-stimulated cells or from cells treated with
   the persulfidating agent Cys persulfide (Cys-SSH) ([110]Fig. 2D). Using
   Cys-SSH standard solutions, we then calculated the persulfide content
   to be 3.25 ± 1.2 nmol/mg protein in the LPS/IFN-γ samples. It was
   previously reported that total thiol content in these cells is
   ∼180 nmol/mg [[111]35,[112]47]; therefore, from these results we
   estimate that in the activated TDM nearly 2% of thiols have undergone
   persulfidation. The level of persulfidation in LPS/IFN-γ-challenged
   cells was comparable to that of Cys–SSH–treated cells ([113]Fig. 2D).
   Using the same methodology in J774 cells, we estimated the persulfide
   content to be 1.1 ± 0.4 nmol/mg and 4.0 ± 0.25 nmol/mg in the control
   and LPS/IFN-γ samples, respectively. It should be considered that the
   Cys-SSH solution used in our experiments is a mixture of several
   species and likely contains some amount of cystine, Cys and Na[2]S (see
   further comments in the Materials and methods section). In control
   experiments we observed that cell exposure to Na[2]S, cystine or Cys
   did not detectibly increase S-persulfidation levels ([114]Supplementary
   Figs. S2A and B), supporting the conclusion that Cys-SSH was
   responsible for the above mentioned effects. However, it remains
   possible that polysulfides present in the Cys-SSH solution [[115]33]
   may partially contribute to the observed effects.

   It is likely that in cells, P-SSH are in equilibrium with low mass
   persulfides such as Cys-SSH and GSSH; however, the nature of this
   equilibrium is not established. Using ultrafiltration fractionation, we
   found that the majority (over 75%) of the RSSH could be ascribed to
   molecules of high mass (>10,000 Da; [116]Fig. 2E). While the
   determination of individual low mass persulfides in macrophages awaits
   further investigation, these results nonetheless suggest that cellular
   RSSH are mostly associated with protein rather than small thiol.

3.3. Large-scale analysis of the macrophage persulfide proteome

   We next set out to identify protein targets of persulfidation in
   resting and LPS/IFN-γ-stimulated cells. For this purpose, we performed
   a proteomic analysis of persulfidation, employing BTA enrichment
   coupled to LC-MS/MS analysis (see Materials and methods for details).
   In parallel, a global proteome analysis was conducted. These
   experiments were performed using THP-1 cells to allow for the
   scalability needed for a proteome-wide analysis. LC-MS/MS analysis of
   three biological replicates resulted in the identification of 3818
   proteins (in the BTA pulldown) across the treatment groups. A total of
   2849 proteins were included in the final analysis following filtration
   for proteins found in at least two out of three replicates in at least
   one experimental group ([117]Table S1). In addition, the proteins in
   the starting lysate were subjected to LC-MS/MS analysis. In the
   whole-cell proteome, a total of 3927 proteins were identified across
   the samples; of these, 3717 proteins were found in at least two out of
   three replicates in at least one experimental group and were retained
   for further analysis ([118]Table S2).

   Next, a more detailed analysis of the persulfide and global proteomes
   was undertaken in order to identify differentially persulfidated or
   expressed proteins (see Materials and methods section), the results of
   which are summarized in the volcano plots shown in [119]Fig. 3A and
   [120]Supplementary Fig. S3A. To determine significant differences in
   protein persulfidation/expression, we applied a threshold of log2
   fold-change >2 and p-value <0.05 (t-test). This analysis demonstrated
   the significant and widespread effect of LPS/IFN-γ stimulation on
   persulfidation, with ∼800 proteins displaying significantly increased
   persulfidation ([121]Fig. 3A). In sharp contrast, we did not detect
   proteins displaying significantly decreased persulfidation. Compared to
   the persulfide proteome, changes in the global proteome were rather
   small, with only 40 proteins identified as upregulated and 14 as
   downregulated in response to LPS/IFN-γ stimulation (log2 fold-change
   >2; p < 0.05) ([122]Supplementary Fig. S3A). These latter results are
   in good agreement with previous studies that analyzed the total
   proteome of M1-activated monocytes/macrophages, particularly
   considering that the experimental conditions as well as the MS analysis
   parameters were not identical among the different studies [[123][48],
   [124][49], [125][50]]. In particular, there is a substantial overlap
   between proteins identified as being upregulated in M1-activated cells
   in our study and the previous studies ([126]Supplementary Fig. S3A).
   Notably, our finding that the redox proteome is more dynamic than the
   total proteome extends recent work by us and others [[127]35,[128]51].

Fig. 3.

   [129]Fig. 3
   [130]Open in a new tab

   Alterations in the persulfide proteome following exposure of
   THP-1 cells to LPS/IFN-γ. (A) Volcano plot visualization of the
   persulfidation profile of 2849 unique proteins following cell treatment
   with LPS/IFN-γ. Proteins with significantly increased persulfidation
   (log2 fold change >2; adjusted p-value <0.05) are highlighted in red.
   (B) Highly enriched biological processes of the LPS/IFN-γ-regulated
   persulfide proteome as determined by Metascape. The categories with
   gene count >6 are shown in ascending order of p-values. (C) Metascape
   visualization of the interactome network of the LPS/IFN-γ-regulated
   persulfide proteome. (For interpretation of the references to color in