Advertisement

SITE-SPECIFIC ACTIVITY-BASED PROTEIN PROFILING USING PHOSPHONATE HANDLES

  • Wouter van Bergen
    Affiliations
    Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Netherlands Proteomics Center, Padualaan 8, Utrecht 3584 CH, The Netherlands
    Search for articles by this author
  • Johannes F. Hevler
    Affiliations
    Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Netherlands Proteomics Center, Padualaan 8, Utrecht 3584 CH, The Netherlands
    Search for articles by this author
  • Wei Wu
    Affiliations
    Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Netherlands Proteomics Center, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Singapore Immunology Network (SigN), Agency for Science, Technology, and Research (A*STAR), 138648 Singapore, Singapore

    Department of Pharmacy, National University of Singapore, 117543, Singapore, Singapore
    Search for articles by this author
  • Marc P. Baggelaar
    Correspondence
    Contact details for correspondence:
    Affiliations
    Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Netherlands Proteomics Center, Padualaan 8, Utrecht 3584 CH, The Netherlands
    Search for articles by this author
  • Albert J.R. Heck
    Correspondence
    Contact details for correspondence: a.j.r.heck.uu.nl
    Affiliations
    Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, Utrecht 3584 CH, The Netherlands

    Netherlands Proteomics Center, Padualaan 8, Utrecht 3584 CH, The Netherlands
    Search for articles by this author
Open AccessPublished:November 23, 2022DOI:https://doi.org/10.1016/j.mcpro.2022.100455

      Highlights

      • Use IMAC-enrichable phosphonate affinity tags (PhosID) for activity-based protein profiling (ABPP) providing directly the ABP-binding sites
      • PhosID–ABPP is compatible with intact cell inhibitor treatment
      • We observed and quantified extensive differences in approachable binding sites in intact cells and in lysates
      • Trypsin and Pepsin provide complementary views of the ABP-interactome

      Abstract

      Most drug molecules target proteins. Identification of the exact drug binding sites on these proteins is essential to understand and predict how drugs affect protein structure and function. To address this challenge, we developed a strategy that uses IMAC-enrichable phosphonate affinity tags, for efficient and selective enrichment of peptides bound to an activity-based probe (ABP), enabling the identification of the exact drug binding site. As a proof of concept, using this approach, termed PhosID–ABPP (Activity-based protein profiling), over 500 unique binding sites were reproducibly identified of an alkynylated Afatinib derivative (PF-06672131). As PhosID–ABPP is compatible with intact cell inhibitor treatment, we investigated the quantitative differences in approachable binding sites in intact cells and in lysates of the same cell line and observed and quantified substantial differences. Moreover, an alternative protease digestion approach was used to capture the previously reported binding site on the epidermal growth factor receptor (EGFR), which turned out to remain elusive when using solely trypsin as protease. Overall, we find that PhosID–ABPP is highly complementary to biotin-based enrichment strategies in activity-based protein profiling studies, with PhosID–ABPP providing the advantage of direct ABP-interaction site identification.

      Graphical abstract

      Keywords

      Abbreviations:

      2-AEP (2-aminoethylphosphonic acid), ABP (Activity-based probe), ABPP (Activity-based protein profiling), ACN (Acetonitrile), AHA (Azidohomoalanine), AIR (Ambiguous interaction restrants), CuAAC (Cu(I)-catalyzed azide–alkyne cycloaddition), DMEM (Dulbecco's modified eagle medium), DMSO (Dimethylsulfoxide), DPBS (Dulbecco's phosphate buffered saline), DTT (Dithiothreitol), EGFR (Epidermal growth factor receptor), ER (Endoplasmic reticulum), FBS (Fetal Bovine Serum), GAP (GTPase-activating protein), GEF (Guanosine exchange factor), GLRX (Glutaredoxin), GSH (Glutathione), GSTP1 (Glutathione S-transferase P), IAA (Iodoacetamide), iBAQ (intensity-based absolute quantification), IMAC (Immobilized metal affinity chromatography), MD (Molecular dynamics), MQ (Milli Q), NCE (Normalized collision energy), NEB (New England BioLabs), NHS (N-Hydroxysuccinimide), NTA (Nitrilotriacetic acid), PANTHER (Protein analysis through evolutionary relationships), PLIP (Protein ligand interaction profiler), PPI (Protein-protein interaction), PRDX (Peroxiredoxin), PSM (Peptide spectrum match), PTM (Post translational modification), RTN4 (Reticulon-4), SDC (Sodium Deoxycholate), TAD (Torsion angle molecular dynamics), TCEP (Tris(2-carboxyethyl)phosphine), THPTA (Tris(3-hydroxypropyltriazolylmethyl)amine), TXN (Thioredoxin)

      Introduction

      Activity-based protein profiling (ABPP) can monitor targets and off-targets of small molecule drugs, and is by now considered a powerful and versatile chemoproteomic strategy to advance drug discovery (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ,
      • Parker C.G.
      • Pratt M.R.
      Click Chemistry in Proteomic Investigations.
      ). ABPP utilizes activity-based probes (ABPs) to interrogate activity or site occupancy status of proteins (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ). The chemical probes used in ABPP generally consist of a ‘warhead’ to form a covalent bond with target proteins, a recognition element that enhances affinity for specific proteins, and a reporter tag that enables visualization or enrichment of targeted proteins (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ,
      • Willems L.I.
      • Overkleeft H.S.
      • Van Kasteren S.I.
      Current developments in activity-based protein profiling.
      ,
      • Serwa R.
      • Tate E.W.
      Activity-Based Profiling for Drug Discovery.
      ,
      • Benns H.J.
      • Wincott C.J.
      • Tate E.W.
      • Child M.A.
      Activity- and reactivity-based proteomics: Recent technological advances and applications in drug discovery.
      ,
      • Van Esbroeck A.C.M.
      • Janssen A.P.A.
      • Cognetta A.B.
      • Ogasawara D.
      • Shpak G.
      • Van Der Kroeg M.
      • Kantae V.
      • Baggelaar M.P.
      • De Vrij F.M.S.
      • Deng H.
      • Allarà M.
      • Fezza F.
      • Lin Z.
      • Van Der Wel T.
      • Soethoudt M.
      • Mock E.D.
      • Den Dulk H.
      • Baak I.L.
      • Florea B.I.
      • Hendriks G.
      • De Petrocellis L.
      • Overkleeft H.S.
      • Hankemeier T.
      • De Zeeuw C.I.
      • Di Marzo V.
      • Maccarrone M.
      • Cravatt B.F.
      • Kushner S.A.
      • Van Der Stelt M.
      Activity-based protein profiling reveals off-target proteins of the FAAH inhibitor BIA 10-2474.
      ,
      • Liu Y.
      • Patricelli M.P.
      • Cravatt B.F.
      Activity-based protein profiling: The serine hydrolases.
      ,
      • Greenbaum D.
      • Baruch A.
      • Hayrapetian L.
      • Darula Z.
      • Burlingame A.
      • Medzihradszky K.F.
      • Bogyo M.
      Chemical Approaches for Functionally Probing the Proteome.
      ,
      • Meissner F.
      • Geddes-McAlister J.
      • Mann M.
      • Bantscheff M.
      The emerging role of mass spectrometry-based proteomics in drug discovery.
      ).
      Protein-protein interactions (PPIs), post-translational modifications (PTMs), and interactions with endogenous small molecules affect protein conformation and activity (
      • Altelaar A.F.M.
      • Munoz J.
      • Heck A.J.R.
      Next-generation proteomics: towards an integrative view of proteome dynamics.
      ,
      • Schastnaya E.
      • Raguz Nakic Z.
      • Gruber C.H.
      • Doubleday P.F.
      • Krishnan A.
      • Johns N.I.
      • Park J.
      • Wang H.H.
      • Sauer U.
      (2021) Extensive regulation of enzyme activity by phosphorylation in Escherichia coli.
      ,
      • Holdgate G.A.
      • Meek T.D.
      • Grimley R.L.
      Mechanistic enzymology in drug discovery: a fresh perspective.
      ,
      • Motlagh H.N.
      • Wrabl J.O.
      • Li J.
      • Hilser V.J.
      The ensemble nature of allostery.
      ). Therefore, it is crucial to monitor drug-protein interactions, ideally in the native protein environment. In contrast with many substrate-based enzyme assays, ABPP allows monitoring drug-protein interactions in complex cellular lysates, cell cultures and can even be used in in vivo studies (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ,
      • Parker C.G.
      • Pratt M.R.
      Click Chemistry in Proteomic Investigations.
      ,
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ).
      Because ABPs generally target a small fraction of the proteins in complex proteomes, enrichment strategies are typically required to identify and quantify ABP targets (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ,
      • Fang H.
      • Peng B.
      • Ong S.Y.
      • Wu Q.
      • Li L.
      • Yao S.Q.
      Recent advances in activity-based probes (ABPs) and affinity-based probes (AfBPs) for profiling of enzymes.
      ,
      • Bird R.E.
      • Lemmel S.A.
      • Yu X.
      • Zhou Q.A.
      Bioorthogonal Chemistry and Its Applications.
      ). Presently, biotin is by far the most popular affinity handle to retrieve ABP-targeted proteins. However, while strong streptavidin-biotin interactions facilitate stringent washing conditions to ensure very selective enrichment, elution from streptavidin is hampered, as harsh conditions are required for release (
      • Schiapparelli L.M.
      • McClatchy D.B.
      • Liu H.-H.
      • Sharma P.
      • Yates J.R.
      • Cline H.T.
      Direct Detection of Biotinylated Proteins by Mass Spectrometry.
      ,
      • Browne C.M.
      • Jiang B.
      • Ficarro S.B.
      • Doctor Z.M.
      • Johnson J.L.
      • Card J.D.
      • Sivakumaren S.C.
      • Alexander W.M.
      • Yaron T.M.
      • Murphy C.J.
      • Kwiatkowski N.P.
      • Zhang T.
      • Cantley L.C.
      • Gray N.S.
      • Marto J.A.
      A Chemoproteomic Strategy for Direct and Proteome-Wide Covalent Inhibitor Target-Site Identification.
      ,
      • Zanon P.R.A.
      • Lewald L.
      • Hacker S.M.
      Isotopically Labeled Desthiobiotin Azide (isoDTB) Tags Enable Global Profiling of the Bacterial Cysteinome.
      ). To circumvent these limitations, on-bead digestion of the bound protein facilitates the detection of the ABP-bound proteins. Unfortunately, this typically does not allow the specific detection of the exact ABP binding site, and ABP-bound protein identification is based on non-ABP-bound peptides, making these assignments somewhat ambiguous.
      Identification of the exact ABP binding sites on proteins with an amino acid specific resolution is desirable because: 1) It will enhance the confidence in identifying bona fide ABP targeted proteins. 2) It yields specific information on drug-protein interactions which can be used as restraints for structural modeling 3) Specific site detection may reveal and distinguish multiple binding sites on the same protein. All in all, site-specific detection of ABPs are critical to improve drug development, as knowledge on binding mechanisms can be used to optimize drug binding and action.
      To address issues with biotin for ABP binding site identification, various alternative approaches have already been explored to identify the ABP-labeled peptides, such as cleavable linkers and the use of desthiobiotin (
      • Browne C.M.
      • Jiang B.
      • Ficarro S.B.
      • Doctor Z.M.
      • Johnson J.L.
      • Card J.D.
      • Sivakumaren S.C.
      • Alexander W.M.
      • Yaron T.M.
      • Murphy C.J.
      • Kwiatkowski N.P.
      • Zhang T.
      • Cantley L.C.
      • Gray N.S.
      • Marto J.A.
      A Chemoproteomic Strategy for Direct and Proteome-Wide Covalent Inhibitor Target-Site Identification.
      ,
      • Zanon P.R.A.
      • Lewald L.
      • Hacker S.M.
      Isotopically Labeled Desthiobiotin Azide (isoDTB) Tags Enable Global Profiling of the Bacterial Cysteinome.
      ,
      • Weerapana E.
      • Wang C.
      • Simon G.M.
      • Richter F.
      • Khare S.
      • Dillon M.B.D.
      • Bachovchin D.A.
      • Mowen K.
      • Baker D.
      • Cravatt B.F.
      Quantitative reactivity profiling predicts functional cysteines in proteomes.
      ,
      • Weerapana E.
      • Speers A.E.
      • Cravatt B.F.
      Tandem orthogonal proteolysis-activity-based protein profiling (TOP-ABPP)—a general method for mapping sites of probe modification in proteomes.
      ). However, it still remains challenging to enrich and detect these low abundant ABP-bound peptides from a complex sample (
      • Weerapana E.
      • Speers A.E.
      • Cravatt B.F.
      Tandem orthogonal proteolysis-activity-based protein profiling (TOP-ABPP)—a general method for mapping sites of probe modification in proteomes.
      ).
      Recently, we reported strategies that employ phosphonate handles to enrich cross-linked peptides (Phox) and clickable phosphonate handles to enrich newly synthesized proteins (PhosID) (
      • Post H.
      • Penning R.
      • Fitzpatrick M.A.
      • Garrigues L.B.
      • Wu W.
      • Macgillavry H.D.
      • Hoogenraad C.C.
      • Heck A.J.R.
      • Altelaar A.F.M.
      Robust, Sensitive, and Automated Phosphopeptide Enrichment Optimized for Low Sample Amounts Applied to Primary Hippocampal Neurons.
      ,
      • Kleinpenning F.
      • Steigenberger B.
      • Wu W.
      • Heck A.J.R.
      (2020) Fishing for newly synthesized proteins with phosphonate-handles.
      ). This latter strategy avoids many of the limitations encountered with biotin, since the IMAC (immobilized metal affinity chromatography) enrichment of chemically stable phosphonate-labeled peptides is highly efficient, readily automated, and facilitates easy release under mild conditions (
      • Post H.
      • Penning R.
      • Fitzpatrick M.A.
      • Garrigues L.B.
      • Wu W.
      • Macgillavry H.D.
      • Hoogenraad C.C.
      • Heck A.J.R.
      • Altelaar A.F.M.
      Robust, Sensitive, and Automated Phosphopeptide Enrichment Optimized for Low Sample Amounts Applied to Primary Hippocampal Neurons.
      ,
      • Kleinpenning F.
      • Steigenberger B.
      • Wu W.
      • Heck A.J.R.
      (2020) Fishing for newly synthesized proteins with phosphonate-handles.
      ). Encouraged by previous studies on direct ABP binding site identification and PhosID, we envisioned that the application of phosphonate-based clickable affinity handles could represent a powerful alternative strategy for ABPP, combining high sensitivity and efficiency with potentially unprecedented performance in exact binding site identification.
      Here we report the application of phosphonate affinity handles for ABPP, enabling exact ABP binding site identification, which we term PhosID–ABPP (Figure 1). In a proof-of-principal ABP study we used the alkynylated Afatinib derivative (PF-06672131), a cysteine reactive ABP known to target the ATP binding pocket of the epidermal growth factor receptor (EGFR) (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ). This probe has been employed in several biotin-based ABPP studies (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ,
      • Shindo N.
      • Fuchida H.
      • Sato M.
      • Watari K.
      • Shibata T.
      • Kuwata K.
      • Miura C.
      • Okamoto K.
      • Hatsuyama Y.
      • Tokunaga K.
      • Sakamoto S.
      • Morimoto S.
      • Abe Y.
      • Shiroishi M.
      • Caaveiro J.M.M.
      • Ueda T.
      • Tamura T.
      • Matsunaga N.
      • Nakao T.
      • Koyanagi S.
      • Ohdo S.
      • Yamaguchi Y.
      • Hamachi I.
      • Ono M.
      • Ojida A.
      Selective and reversible modification of kinase cysteines with chlorofluoroacetamides.
      ), and revealed multiple off-targets for which the ABP binding sites were unknown.
      Figure thumbnail gr1
      Figure 1Schematic representation of the here developed chemoproteomic site identification PhosID-ABPP strategy. (
      • Cravatt B.F.
      • Wright A.T.
      • Kozarich J.W.
      Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
      ) Incubation of the activity-based probe (ABP) PF-06672131 in intact cells or in cell lysates. (
      • Parker C.G.
      • Pratt M.R.
      Click Chemistry in Proteomic Investigations.
      ) Cell lysis and attachment of the phosphonate handle by Cu(I)-catalyzed azide–alkyne cycloaddition (CuAAC). (
      • Willems L.I.
      • Overkleeft H.S.
      • Van Kasteren S.I.
      Current developments in activity-based protein profiling.
      ) Subsequent, enzymatic digestion by trypsin or pepsin. (
      • Serwa R.
      • Tate E.W.
      Activity-Based Profiling for Drug Discovery.
      ) Dephosphorylation to remove endogenous phosphorylation. (
      • Benns H.J.
      • Wincott C.J.
      • Tate E.W.
      • Child M.A.
      Activity- and reactivity-based proteomics: Recent technological advances and applications in drug discovery.
      ) Automated Fe3+-IMAC enrichment of ABP-labeled peptides. (
      • Van Esbroeck A.C.M.
      • Janssen A.P.A.
      • Cognetta A.B.
      • Ogasawara D.
      • Shpak G.
      • Van Der Kroeg M.
      • Kantae V.
      • Baggelaar M.P.
      • De Vrij F.M.S.
      • Deng H.
      • Allarà M.
      • Fezza F.
      • Lin Z.
      • Van Der Wel T.
      • Soethoudt M.
      • Mock E.D.
      • Den Dulk H.
      • Baak I.L.
      • Florea B.I.
      • Hendriks G.
      • De Petrocellis L.
      • Overkleeft H.S.
      • Hankemeier T.
      • De Zeeuw C.I.
      • Di Marzo V.
      • Maccarrone M.
      • Cravatt B.F.
      • Kushner S.A.
      • Van Der Stelt M.
      Activity-based protein profiling reveals off-target proteins of the FAAH inhibitor BIA 10-2474.
      ) Identification of ABP labeled peptides that are separated, sequenced and identified by LC-MS/MS. (
      • Liu Y.
      • Patricelli M.P.
      • Cravatt B.F.
      Activity-based protein profiling: The serine hydrolases.
      ) Mass triggering to ensure extensive fragmentation of ABP-labeled peptides.

      Experimental procedures

      Synthesis of the phosphonate-azide

      A stock solution of 500 mM 2-aminoethyl phosphonic acid (2-AEP, Sigma-Aldrich) was prepared in 1× Milli Q water (MQ, Millipore) and adjusted to pH 10 using sodium hydroxide. 75 μL of 0.5 M 2-AEP (Sigma-Aldrich) was incubated with 20 μL of 400 mM azidobutyric acid NHS (N-Hydroxysuccinimide) ester (Lumiprobe) in dimethylsulfoxide (DMSO, Sigma-Aldrich) and 45 μL MQ. The reaction was incubated for at least 2 h at room temperature in the dark rotating, resulting in approximately 50 mM stocks of phosphonate-azide.

      Cell culture

      A431 cells (CRL-1555, ATCC) with a passage number below 20 were cultured growth medium (Dulbecco’s modified eagle medium (DMEM, Lonza) supplemented with Fetal Bovine Serum (FBS, HyClone GE) and 1X L-Glutamine (Lonza)). Cells were kept in a humidified atmosphere with 5% CO2 at 37°C in T175 flasks (Greiner). A431 cells were split twice a week by washing with Dulbecco’s phosphate buffered saline (DPBS, Lonza) and treatment with 0.05% Trypsin-EDTA (Gibco) for cell detachment. After detachment, trypsin was quenched by adding growth medium. 1/20 of the cell suspension was taken and grown with fresh growth medium in a new T175 flask.

      PF-06672131 incubation in intact cells

      5e6 A431 cells were plated in 15cm plates (Greiner) 24 h before probe treatment and kept in a humidified atmosphere with 5% CO2 at 37°C. The growth medium was replaced by treatment medium (growth medium with 25 μM PF-06672131 (Sigma-Aldrich)) and incubated at 37°C, 5% CO2 for 4h. Cells were detached with 0.25% Trypsin-EDTA (Gibco) and the cell suspension was spun down at 400 g for 5 minutes and the supernatant was aspirated. The cell pellet was washed with DPBS before snap freezing the cell pellet in liquid nitrogen. The cell pellet was stored at −80°C for later use.

      Cell lysis

      Cell pellets were lysed in 500 μL 1% sodium deoxycholate (SDC, Sigma), 1X Protease inhibitor cocktail EDTA (Roche) in DPBS (Lonza) per 15 cm plate. The cells were suspended and incubated on ice for 30 minutes. Sonication was performed with a UP100H probe tip sonicator (Hielscher) using 60% amplitude with 1 seconds on and 1 seconds off for 1 minute. Cell debris and DNA was spun down for 30 minutes at 20567 g at 16°C. The supernatant was collected and the protein concentration was determined by a bicinchoninic acid assay (Thermo Fisher Scientific).

      PF-06672131 incubation in cell lysate

      5.0 mg A431 cell lysates were treated with 100 μM PF-06672131 in 1 mL for 1 h at 37°C. Methanol-chloroform precipitation was performed and the air-dried pellets were resuspended in 500 μL 8M urea (Sigma-Aldrich). The samples were sonicated with a bioruptor (Diagenode) for 10 minutes with 30 seconds on and 30 seconds off at high amplitude to fully dissolve all proteins.

      Sample processing for SDS-PAGE

      Cu(I)-catalyzed azide–alkyne cycloaddition (CuAAC) was performed on 30 μg protein lysates in 1× DPBS (pH 7.5) in 20 μL total volume. 2 μL mastermix of CuAAC components was added to a final concentrations of: 15 μM of azide-fluor 488, 150 μM tris(3-hydroxypropyltriazolylmethyl)amine (THPTA; Lumiprobe), 1.5 mM CuSO4 5·H2O (Sigma-Aldrich), and 1.5 mM tris(2-carboxyethyl)phosphine (TCEP, Sigma-Aldrich). Samples were incubated for 2 h at room temperature while rotating. 200 mM Dithiothreitol (DTT, Sigma) in 4X sample buffer XT (Bio-Rad) was added to reach a final concentration of 50 mM DTT and 1X sample buffer. The samples were boiled at 95°C for 5 minutes. Afterwards, the samples were cooled down to room temperature and loaded on a 4-12% bis-tris protein gel (Bio-Rad). Precision Plus Protein Dual Color Standards (Bio-Rad) was used as a molecular weight marker. After running, the gel was scanned with the “Cy2 channel” using an Amersham Imager 600 (GE Healthcare) to visualize the azide-fluor 488. Finally, Imperial blue protein stain (Thermo Fisher Scientific) was used to visualize the total protein loaded.

      Bioorthogonal chemistry reactions for proteomics

      CuAAC was performed on 5.0 mg protein lysates in 2 M urea (Merck) in 1× DPBS (pH 7.5). CuAAC components were added in the following order: 5 mM THPTA, 2.5 mM CuSO4 5·H2O, 500 μM phosphonate-azide (for preparation see “Synthesis of the phosphonate-azide”) and 25 mM sodium ascorbate (Sigma-Aldrich) in a final volume of 2 mL. Samples were incubated for 2h at room temperature while rotating. Methanol-chloroform precipitation was performed to remove the CuAAC components and the air-dried pellets were resuspended in 500 μL 8M urea and sonicated in a bioruptor with high amplitude for 10 minutes with cycles of 30 seconds on and 30 seconds off.

      Sample processing for digestion

      Clicked and dissolved protein samples were diluted to 4M urea with 50 mM ammonium bicarbonate (pH 8, AmBic, Sigma-Aldrich). The proteins were reduced with 4 mM dithiothreitol (DTT, Sigma-Aldrich) for 60 minutes at room temperature and alkylated in the dark using 8 mM iodoacetamide (IAA, Sigma-Aldrich) for 30 min. Residual IAA was quenched by adding DTT to a final concentration of 4 mM. Next, samples were diluted 2X with 50 mM AmBic and digested with LysC (1:75 enzyme to protein ratio, Wako) for 4 h at 37 °C. Finally, proteins were digested overnight using Trypsin (1:50, enzyme to protein ratio, Sigma-Aldrich) at 37 °C in a final volume of 2 mL. Digested material was desalted using 3cc C18 Seppak cartridges (Waters) and air dried using a vacuum centrifuge.
      For the digestion with pepsin, protease incubation (Porcine, 1:50, enzyme to protein ratio, Sigma-Aldrich) was performed for 4 h at 37 °C in 40 mM HCl in a total volume of 2 mL (pH 2). After incubation, pepsin was irreversible inactivated by adjusting the pH > 6 with 1M AmBic. Digested material was desalted using 3cc C18 Seppak cartridges and air dried using a vacuum centrifuge.

      Dephosphorylation

      Samples were dephosphorylated prior to IMAC enrichment. Desalted peptides were dissolved in 1 mL 1× CutSmart buffer (New England BioLabs, NEB) and incubated with 50 units Alkaline Phosphatase (calf intestinal, QuickCIP, NEB) overnight at 37 °C while shaking. We previously showed that in contrast to normal phosphopeptides, the peptides modified with the probe-phosphonate-handles are unaffected by phosphatase treatment (
      • Kleinpenning F.
      • Steigenberger B.
      • Wu W.
      • Heck A.J.R.
      (2020) Fishing for newly synthesized proteins with phosphonate-handles.
      ). Following the dephosphorylation, all peptides were again desalted using 3cc C18 Seppak cartridges (Waters) and air dried using a vacuum centrifuge.

      Automated Fe3+-IMAC enrichment

      Probe-phosphonate-labeled peptides were enriched using Fe(III)-NTA 5 μL (Agilent technologies) in an automated fashion by the AssayMAP Bravo Platform (Agilent Technologies). Fe(III)-NTA (nitrilotriacetic acid) cartridges were primed at a flow rate of 100 μL/min with 250 μL of priming buffer (0.1% TFA, 99.9% ACN) and equilibrated at a flow rate of 50 μL/min with 250 μL of loading buffer (0.1% TFA, 80% ACN). The flow through was collected into a separate plate. Dried peptides were dissolved in 200 μL of loading buffer and loaded at a flow rate of 5 μL/min onto the cartridge. Columns were washed with 250 μL of loading buffer at a flow rate of 20 μL/min, and the phosphonate-labeled peptides were eluted with 35 μL of ammonia (10%) at a flow rate of 5 μL/min directly into 35 μL of formic acid (10%). Flowthroughs and elutions were air dried afterwards and injected directly on a liquid chromatography-coupled mass spectrometer.

      LC-MS/MS

      Samples were analyzed on a nanospray UHPLC system Ultimate3000 (Thermo Fisher Scientific) coupled to an Orbitrap Exploris 480 mass spectrometer (Thermo Fisher Scientific), in data-dependent acquisition mode. Peptides were trapped on an Acclaim Pepmap 100 C18 (5mm x 0.3 mm, 5 μm) in solvent A (0.1 % v/v formic acid in water) and then separated on an analytical column (Poroshell 120 EC C18, 50 cm x 75μm , 2.7 μm, Agilent Technologies) with a flowrate of 300 nL/min. Elution fractions were measured with a gradient 9% solvent B (0.1% v/v formic acid in 80% acetonitrile (ACN)) for 1 min, 13–45% in 37 min, 45–99% in 3 min, and 99% solvent B for 4 min was used. Finally, the system was equilibrated back to 91% solvent A for 8 min. Periodic MS1 scans were performed at a resolution of 60,000, between 375 and 2000 m/z after reaching the normalized AGC target with automatic injection time every second. Top intense precursors were fragmented with normalized collision energy (NCE) of 28% and 12 s dynamic exclusion time in between 2 MS1 scans. HCD fragmentation was performed on precursors at a resolution of 30,000. Extra stepped-HCD scans with 28, 32 and 36% NCE were performed on a precursor each time 194.05822 or 247.08477 m/z ions were detected in the first MS2 scan, being signature fragment ions indicative of the ABP-probe.
      Total proteomes (before enrichment) were measured with a gradient of 9% solvent B for 1 min, 13–44% in 97 min, 44–99% in 3 min, and 99% solvent B for 4 min was used. Finally, the system was equilibrated back to 91% solvent A for 10 min. Periodic MS1 scans were performed at a resolution of 60,000, between 375 and 1600 m/z after reaching the normalized AGC target with automatic injection time every second. Top intense precursors were fragmented with normalized collision energy (NCE) of 28% and 16 s dynamic exclusion time in between 2 MS1 scans.

      Database search and analysis

      Trypsin-digested LC-MS/MS run files were processed using MaxQuant 2.0.1.0 or higher and the Andromeda search engine and searched against the human Swissprot database (version September 2020, 20375 entries) (
      • Cox J.
      • Mann M.
      MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.
      ). Enzyme specificity was set to Trypsin (C-terminal cleavage of lysine or arginine, except when proline follows), and up to 3 missed cleavages were allowed. Minimum peptide length was set to 4. Variable modifications of cysteine carbamidomethylation, methionine oxidation, carbamylation of lysines and N-termini, protein N-terminal acetylation and phosphorylation on serine, threonine and tyrosine were allowed, together with PF-06672131-phosphonate adduct on cysteine (C29H34N9O6ClFP; 689.20422). Diagnostic ions (194.05822, 247.08477, 690.21205 and 416.12896) and neutral losses (−274.08309, −292.09366, and −689.20422) were annotated as specific characteristic fragment ions for PF-06672131-phosphonate. Mass tolerance for precursors and fragment ions was 4.5 and 20 ppm respectively. The integrated MaxQuant contaminants database was used to filter out for contaminants. A false discovery rate (FDR) of 1% for PSMs and proteins was allowed using a target-decoy approach. A score cutoff of 40 was used for modified peptides. For total proteome measurements, iBAQ was enabled.
      Pepsin digested LC-MS/MS run files were searched against the human (20375 entries) Swissprot database (version September 2020) using Fragpipe v18.0 with the MSFragger 3.5 and philosopher 4.3.0 search engine using the default settings (
      • Kong A.T.
      • Leprevost F.V.
      • Avtonomov D.M.
      • Mellacheruvu D.
      • Nesvizhskii A.I.
      MSFragger: Ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics.
      ). The integrated Fragpipe contaminant database was used for filtering out contaminants. Cleavage site was set to nonspecific and a peptide length between 6 and 30 was allowed. Oxidation of methionine, acetylation of the protein N-terminus, carbamidomethylation of cysteines were set as variable modifications. PF-06672131-phosphonate (689.20422) adduct was also set as a variable modification on cysteine. Precursor and fragment mass tolerance were set to 20 ppm both. FDR for PSMs and proteins was set to 1% using a target-decoy approach.

      Statistical analysis and visualization

      For MaxQuant output, the tables “evidence.txt” and “PF-06672131-phosphonateSites.txt” were used to generate a comprehensive table of the ABP binding sites and their relative abundance. Reverse and potential contaminants were filtered out. Intensities were used to calculate the relative abundance of modification in each sample. To calculate the relative abundance of proteins in A431 cells, the average iBAQ values were taken from the “proteinGroups.txt” table. For analysis of the ABP binding sites in IMAC elutions, only peptides modified with PF-06672131-phosphonate were kept for ABP binding site analysis. Peptides that were found in 2 out of 3 replicates were considered as ABP binding sites. Peptides with ambiguous localization of PF-06672131 (localization probability<0.75) were used for assessing the total numbers and intensities of PF-06672131-bound peptides, but not for site specific analysis. Raw peptide intensities of ABP-bound peptides were log2 transformed. Data was checked for normal distribution before performing imputation of missing data with a downshift of 1.8 standard deviations and performing a student’s t-test. For samples digested with pepsin, the “psm.tsv” table was used for analysis, a minimal of 2 (peptide-spectrum match) PSMs per ABP binding site was accepted as a PF-06672131 binding site. Peptides with ambiguous localization of PF-06672131 (i.e. peptides with multiple cysteines) were used for assessing the total number of PF-06672131-bound peptides, but not for site specific analysis. Analysis and visualization of data was done with Perseus 1.6.15, Excel 2016 and Graphpad Prism 9 (
      • Tyanova S.
      • Temu T.
      • Sinitcyn P.
      • Carlson A.
      • Hein M.Y.
      • Geiger T.
      • Mann M.
      • Cox J.
      The Perseus computational platform for comprehensive analysis of (prote)omics data.
      ). Venn diagrams were created using Biovenn (
      • Hulsen T.
      • de Vlieg J.
      • Alkema W.
      BioVenn - A web application for the comparison and visualization of biological lists using area-proportional Venn diagrams.
      ). MS/MS spectra were visualized using in-house software and figures were finalized in Adobe Illustrator.

      Experiment Design and Statistical Rationale

      Lysate- and intact cell-treated samples combined with trypsin-based proteolysis were conducted with n = 3 biological replicates, enriched and injected separately into the LC-MS/MS system. Each raw file was separately processed using the MaxQuant software. This number was sufficient to evaluate reproducibility and quantitatively compare the two conditions. Pepsin-based digestion was conducted with n = 1 replicate on an intact cell-treated sample, which is sufficient for proof of principle with an alternative protease without quantitative analysis.

      Protein - Ligand docking

      A crystal structure for RHOA bound to GDP (1FTN) was prepared for docking using PDB tools (
      • Solca F.
      • Dahl G.
      • Zoephel A.
      • Bader G.
      • Sanderson M.
      • Klein C.
      • Kraemer O.
      • Himmelsbach F.
      • Haaksma E.
      • Adolf G.R.
      Target binding properties and cellular activity of afatinib (BIBW 2992), an irreversible ErbB family blocker.
      ,
      • Rodrigues J.P.G.L.M.
      • Teixeira J.M.C.
      • Trellet M.
      • Bonvin A.M.J.J.
      pdb-tools: a swiss army knife for molecular structures.
      ,
      • Jiménez-García B.
      • Teixeira J.M.C.
      • Trellet M.
      • Rodrigues J.P.G.L.M.
      • Bonvin A.M.J.J.
      PDB-tools web: A user-friendly interface for the manipulation of PDB files.
      ,
      • Honorato R.V.
      • Koukos P.I.
      • Jiménez-García B.
      • Tsaregorodtsev A.
      • Verlato M.
      • Giachetti A.
      • Rosato A.
      • Bonvin A.M.J.J.
      Structural Biology in the Clouds: The WeNMR-EOSC Ecosystem.
      ,
      • Wei Y.
      • Zhang Y.
      • Derewenda U.
      • Liu X.
      • Minor W.
      • Nakamoto R.K.
      • Somlyo A.V.
      • Somlyo A.P.
      • Derewenda Z.S.
      Crystal structure of RhoA–GDP and its functional implications.
      ). A distance restraint of 1.7-1.9 Ångström between the sulfur on the targeted cysteine and the carbon on the probe was set as an unambiguous restraint. Residues (13-20, 33-37, 60-62, 118, 121, and 160-162) in the GDP binding pocket were provided as ambiguous restraint, only used in the rigid body docking protocol. Docking was performed using the default protocol with minor adjustments for small molecule docking in HADDOCK 2.4 (
      • Honorato R.V.
      • Koukos P.I.
      • Jiménez-García B.
      • Tsaregorodtsev A.
      • Verlato M.
      • Giachetti A.
      • Rosato A.
      • Bonvin A.M.J.J.
      Structural Biology in the Clouds: The WeNMR-EOSC Ecosystem.
      ,
      • Van Zundert G.C.P.
      • Rodrigues J.P.G.L.M.
      • Trellet M.
      • Schmitz C.
      • Kastritis P.L.
      • Karaca E.
      • Melquiond A.S.J.
      • Van Dijk M.
      • De Vries S.J.
      • Bonvin A.M.J.J.
      The HADDOCK2.2 Web Server: User-Friendly Integrative Modeling of Biomolecular Complexes.
      ). Rigid body docking was performed using 10,000 structures and the best 400 structures were selected for semi-flexible refinement according to ambiguous interaction restraints (AIR) energies. The resulting 400 structures were analyzed and clustered according to RMSD with a cutoff of 2 and minimal cluster size of 4. The scoring for Evdw in the rigid body docking phase was set to 1.0 and the scoring for Eelec in the water refinement stage was set to 0.1. Moreover, number of molecular dynamics (MD) steps for rigid body and first rigid body cooling stage torsion angle molecular dynamics (TAD) were set to 0. The initial temperature for second and third TAD cooling and were set at 500 and 300 respectively. The best structure for RHOA was chosen based on the lowest HADDOCK scores, distance between the sulfur atom on the targeted Cys16 and the carbon atom on PF-06672131, and visual examination. Protein ligand interaction profiler (PLIP) web server was used to analyze the interactions between the probe and protein in the generated model (
      • Adasme M.F.
      • Linnemann K.L.
      • Bolz S.N.
      • Kaiser F.
      • Salentin S.
      • Haupt V.J.
      • Schroeder M.
      PLIP 2021: expanding the scope of the protein–ligand interaction profiler to DNA and RNA.
      ). 3D models were visualized and exported using ChimeraX or PyMOL (
      • Pettersen E.F.
      • Goddard T.D.
      • Huang C.C.
      • Meng E.C.
      • Couch G.S.
      • Croll T.I.
      • Morris J.H.
      • Ferrin T.E.
      UCSF ChimeraX: Structure visualization for researchers, educators, and developers.
      ).

      Results and discussion

      PF-06672131 protein labeling in cell lysates and intact cells analyzed by in-gel fluorescence

      To assess the efficiency of protein labeling by the probe, we first incubated an A431 (human skin cancer cell line) cell lysate and intact A431 cells with the ABP PF-06672131. After labeling, an azide functionalized fluorescent reporter was introduced using the biorthogonal (CuAAC) reaction. In gel fluorescence revealed that the probe labeled many proteins in both cell lysates and intact cells, indicating that the used ABP displays a broad reactivity across the proteome (Supplementary Figure 1), as was reported previously (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ). This reactivity pattern is similar in lysates and intact cells. However, we also observed differential labeling between cell lysate and intact cells (Supplementary Figure 1). Therefore, we next sought to identify the differences in binding sites of PF-06672131 in cell lysates and intact cells with PhosID.

      Identification of probe-modified peptides by MS/MS fragmentation

      Detailed knowledge of the fragmentation pattern of ABP-bound peptides is crucial for their identification and detection, therefore initially manual examination of MS/MS spectra of ABP-bound peptides was conducted (
      • Ficarro S.B.
      • Browne C.M.
      • Card J.D.
      • Alexander W.M.
      • Zhang T.
      • Park E.
      • Mcnally R.
      • Dhe-Paganon S.
      • Seo H.-S.
      • Lamberto I.
      • Eck M.J.
      • Buhrlage S.J.
      • Gray N.S.
      • Marto J.A.
      Leveraging Gas-Phase Fragmentation Pathways for Improved Identification and Selective Detection of Targets Modified by Covalent Probes.
      ,
      • Renuse S.
      • Madugundu A.K.
      • Jung J.H.
      • Byeon S.K.
      • Goldschmidt H.L.
      • Tahir R.
      • Meyers D.
      • Kim D.I.
      • Cutler J.
      • Kim K.P.
      • Wu X.
      • Huganir R.L.
      • Pandey A.
      Signature Fragment Ions of Biotinylated Peptides.
      ). At first glance, when compared to unmodified peptides, the MS/MS spectra revealed multiple abundant non-annotated high-intensity peaks. We found that many of these initially non-annotated peaks in the MS/MS spectra originated from distinct neutral losses, namely 274.08 and 292.09 m/z (Figure 2A), consistent with fragmentation at the ether bond (Figure 2C). In addition, we observed the loss of the complete ABP-phosphonate (−689.20 m/z). Also, diagnostic ions at 194.06 m/z corresponding to fragmentation of the phosphonate moiety at the triazole ring could be robustly detected in the MS/MS spectra (Figure 2A and B), together with other PF-06672131-specific diagnostic ions (247.08, 416.13, 645.15, 690.21, and 724.20 m/z). Moreover, unreacted probe-phosphonate and fragments thereof are even observed at the MS1 level (690.21, 663.13, 645.15, and 416.13 m/z; Supplementary Figure 2A). By adding the neutral losses and the diagnostic ions in our MS/MS search strategy we could increase the identification success of ABP-labeled peptides by 75% and increase the intensity of ABP-labeled peptides by 54% on average (Figure 2C).
      Figure thumbnail gr2
      Figure 2Selective enrichment and identification of PF-06672131 binding sites. A. MS/MS spectrum of a PF-06672131-phosphonate-bound peptide originating from Reticulon-4 (RTN4) labeled at position 1101. a: a-ion, b: b-ion, y: y-ion, NL: Neutral loss, DI: Diagnostic ion. B. Fragmentation sites observed in the phosphonate-clicked ABP attached to peptides. Diagnostic ions originate from PF-06672131-phosphonate (194.06, 247.08, 416.13, 645.15, 690.21, and 724.20 m/z). The diagnostic ions with an m/z of 194.06 and 247.08 (orange) were used for mass triggering. The neutral losses −274.09, −292.09, −689.20 m/z are indicated in green. C. Inclusion of neutral losses and diagnostic ions originating from the ABP-phosphonate modification in the MaxQuant database search boosts the identification of ABP-bound peptides. D. Enrichment efficiency of intact cell treated samples. Total MS1 intensity of peptides detected in the LC-MS/MS runs before (input) and after (elution) enrichment. E. Correlation of the relative abundance of ABP bound peptides detected by LC-MS/MS runs of lysed- and intact cell-treated samples. The correlation graphs of all individual replicates are provided in . F/G. Venn diagrams showing the overlap of detected ABP-labeled peptides between replicates in lysate (F), and intact cells (G). H. Indication of ABP-labeled proteins (green) in an S-plot of proteins in A431 cells ranked by their relative abundance as estimated from their iBAQ values before enrichment. Proteins that are further discussed are indicated by the largest dark green dots.

      Phosphonate handles enable enrichment of probe-bound peptides

      As in most ABP experiments, peptides labeled by the activity-based probe are relatively low in abundance compared to the background of the unmodified peptides. In our experiments, only 0.001% of the relative peptide abundance originated from ABP-labeled peptides (Figure 2D; Supplementary Figure 2C). The low abundance of ABP-labeled peptides presents a significant additional challenge compared to the previously published PhosID method, in which the azidohomoalanine (AHA) labeled peptides constitute 0.2% of the total peptides quantified (
      • Kleinpenning F.
      • Steigenberger B.
      • Wu W.
      • Heck A.J.R.
      (2020) Fishing for newly synthesized proteins with phosphonate-handles.
      ), or regular phosphoproteomics, in which phosphopeptides generally constitute 2 – 3% of the total peptide intensity before enrichment (
      • Manning G.
      • Plowman G.D.
      • Hunter T.
      • Sudarsanam S.
      Evolution of protein kinase signaling from yeast to man.
      ,
      • Moorhead G.B.G.
      • De Wever V.
      • Templeton G.
      • Kerk D.
      Evolution of protein phosphatases in plants and animals.
      ,
      • Liu X.
      • Rossio V.
      • Thakurta S.G.
      • Flora A.
      • Foster L.
      • Bomgarden R.D.
      • Gygi S.P.
      • Paulo J.A.
      Fe3+-NTA magnetic beads as an alternative to spin column-based phosphopeptide enrichment.
      ).
      Consistent with the relative low abundance of ABP labeled peptides, we identified, without using any enrichment, only 2 peptide spectrum matches (PSMs) for probe labeled peptides in A431 cell lysates. In contrast, by using the efficient automated phosphorylated peptide enrichment on an Assaymap BRAVO system using High-Capacity Fe(III)-NTA Cartridges, we could increase the relative abundance of activity-based probe labeled peptides on average to about 23%, identifying around 1500 PSMs (i.e. a 750-fold increase in PSMs). After the IMAC-based enrichment the remaining 77% of intensity constituted of phosphopeptides (14%) and unmodified peptides (63%) (Figure 2D, Supplementary Figure 2C and D). On the MS1 chromatography trace, the change in sample constitution before and after enrichment shows a depletion of a majority of the (unmodified) peptides (Supplementary Figure 2A).
      The use of the probe-related low mass diagnostic ions in the MS/MS spectra turned out to be a powerful tool to increase the sensitivity of our method (
      • Sweet S.M.M.
      • Creese A.J.
      • Cooper H.J.
      Strategy for the Identification of Sites of Phosphorylation in Proteins: Neutral Loss Triggered Electron Capture Dissociation.
      ). Therefore, in addition to the offline phosphopeptide enrichment, we used a mass triggered method to increase the sensitivity aiming to identify also lower abundant probe labeled peptides. We therefore used probe-specific diagnostic fragment ions of 194.06 and 247.08 m/z.

      PhosID-ABPP reproducibly identifies ABP binding sites in either lysates or intact cells

      Using this optimized strategy, we were able to detect more than 500 ABP binding sites in at least 2 out of the 3 experiments performed on A431 lysates (Figure 2F, Supplementary Data 1). To investigate if we could extend our strategy to detect ABP binding sites in intact cells, we incubated intact A431 cells with PF-06672131. Our method performed also very well when applied to intact cells and we reproducibly identified, under these conditions, around 600 ABP binding sites (Figure 2G, Supplementary Data 1). Interestingly, we also found that the ABP labeled multiple unique binding sites in many proteins. (Supplementary Figure 2E). To validate if the ABP binding site detection strategy does not affect the ABP target scope, we compared the protein targets that were detected earlier by Lanning et al. who used a biotin-based protein-centric ABPP approach to find PF-06672131 protein targets (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ). 302 of the 437 proteins that we identified in intact A431 cells were also found by Lanning et al. by intact cell treatment with PF-06672131. Additionally, we observed 43 probe binding sites on 25 kinases, of which 11 were also detected by Lanning et al. The high overlap indicates that the ABP binding sites that we identify by PhosID–ABPP are within proteins that are enriched by using the earlier described biotin-based affinity approach (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ). From this, we conclude that the PhosID–ABPP approach does not seriously affect the target landscape of the ABP.
      To assess the relative abundance of ABP-targeted proteins in the A431 cells, we assessed the protein abundance in A431 cells as determined by intensity-based absolute quantification (iBAQ) in the LC-MS measurements before Fe3+-IMAC enrichment (Figure 2H, Supplementary Data 2). This comparison revealed that many abundant proteins within the proteome were found to interact with the probe, which may be caused by, in contrast with less abundant proteins, low labeling stoichiometry of abundant proteins being already sufficient for ABP-labeling site detection. Importantly, PhosID–ABPP was found to detect binding sites on proteins ranging deep into the lower abundant regions of the proteome (Figure 2H).

      PhosID–ABPP exposes protein and site-specific differences in protein binding when applied to lysates or intact cells

      We observed remarkable differences in ABP labeling between intact cells and cell lysates (Figure 3B). Nonetheless, the correlation between the intensities of ABP-bound peptides in replicate experiments was found to be high, both for lysate labeled replicates (R2 >0.7) or intact cell labeled replicates (R2 >0.8). However, the correlation between the intensities of ABP-bound peptides intact cells and cell lysates was consistently found to be below an R2 of 0.3 (Figure 2E, Supplementary Figure 3). Moreover, a venn diagram and statistical analysis of the PF-06672131 binding sites observed in the intact cells and the cell lysates showed that there exists an overlap in binding sites, but also a substantial difference in probe binding sites between lysates and intact cells (Figure 3A and B, Supplementary Data 3). The observed difference might be affected by using two different ABP concentrations in cells lysate (25 μM) and intact cells (10 μM), and may also be influenced by the lysis conditions. However, these factors do likely not explain the stark contrast in ABP labeling efficiencies between cell lysates and intact cells for specific sites. To gain insight into the origin of these differences, we focused on some of the most pronounced site specific differences observed focusing on a subset of ABP-targeted proteins.
      Figure thumbnail gr3
      Figure 3Substantial differences observed in ABP-labeling of cell lysates and intact cells. A. Venn diagram showing an overlap of the PF-06672131-bound peptides detected in intact cells (red) and cell lysates (blue). B. Volcano plot showing the difference in labeling intensity through a t-test between lysates and intact cells. ABP binding sites highlighted in red are >2-fold significantly enriched in intact cells compared to lysates, ABP binding sites highlighted in blue are >2-fold significantly enriched in cell lysates compared to intact cells. C. PF-06672131 binds catalytic cysteines on redox-sensitive proteins. Log2 intensities from PF-06672131-bound peptides derived from redox proteins in lysates (blue) and intact cells (red). * Shared tryptic peptide (HGEVCPAGWK) around the peroxidatic cysteine on PRDX1 and PRDX4 was detected next to the unique peptides for PRDX1;Cys173 (HGEVCPAGWKPGSDTIKPDVQK) and PRDX4;Cys245 (HGEVCPAGWKPGSETIIPDAGK).

      Reticulon-4 (RTN4) is a top binding target of the ABP in intact cells

      The labeling of a target protein depends on both the specific reactivity of the probe towards that site/protein and the abundance of this protein in the proteome. Consequently, proteins that are low in abundance in the proteome, but high in abundance in our data set of probe-labeled peptides are likely earnest targets of the probe. Log2 ratios of ABP-peptide intensity over iBAQ values were calculated and used as an approach to prioritize ABP target sites (Supplementary Data 1). A peptide containing probe-labeled Cys1101 of RTN4 belongs to the top 10 most intense ABP labeled peptides in our data on the intact cells. Interestingly, this peptide is labeled around 55-fold more intense in intact cells compared to cell lysates. RTN4 has a relatively low abundance in A431 cells (iBAQ: 4e7, log2(probe-peptide intensity/iBAQ): 5.44, Figure 2H, 3B and Supplementary Data 1, 2 and 3). RTN4 is thought to be located in the endoplasmic reticulum (ER) and plays an important role in maintaining the ER and the formation of ER tubules (
      • Jozsef L.
      • Tashiro K.
      • Kuo A.
      • Park E.
      • Skoura A.
      • Albinsson S.
      • Rivera-Molina F.
      • Harrison K.D.
      • Iwakiri Y.
      • Toomre D.
      • Sessa W.C.
      Reticulon 4 Is Necessary for Endoplasmic Reticulum Tubulation, STIM1-Orai1 Coupling, and Store-operated Calcium Entry.
      ). RTN4 is emerging as a promising target in cancer therapy, and targeting of Cys1101 by the covalent inhibitor DKM 3-30 has been reported to result in aberrant ER tubule formation and mitosis, reducing tumor cell growth (
      • Jozsef L.
      • Tashiro K.
      • Kuo A.
      • Park E.
      • Skoura A.
      • Albinsson S.
      • Rivera-Molina F.
      • Harrison K.D.
      • Iwakiri Y.
      • Toomre D.
      • Sessa W.C.
      Reticulon 4 Is Necessary for Endoplasmic Reticulum Tubulation, STIM1-Orai1 Coupling, and Store-operated Calcium Entry.
      ,
      • Bateman L.A.
      • Nguyen T.B.
      • Roberts A.M.
      • Miyamoto D.K.
      • Ku W.M.
      • Huffman T.R.
      • Petri Y.
      • Heslin M.J.
      • Contreras C.M.
      • Skibola C.F.
      • Olzmann J.A.
      • Nomura D.K.
      Chemoproteomics-enabled covalent ligand screen reveals a cysteine hotspot in reticulon 4 that impairs ER morphology and cancer pathogenicity.
      ,
      • Pathak G.P.
      • Shah R.
      • Kennedy B.E.
      • Murphy J.P.
      • Clements D.
      • Konda P.
      • Giacomantonio M.
      • Xu Z.
      • Schlaepfer I.R.
      • Gujar S.
      (2018) RTN4 Knockdown Dysregulates the AKT Pathway, Destabilizes the Cytoskeleton, and Enhances Paclitaxel-Induced Cytotoxicity in Cancers.
      ). The targeted cysteine is hypothesized to be exposed on the cytosolic side of the ER membrane, making it readily accessible for covalent labeling (
      • Bateman L.A.
      • Nguyen T.B.
      • Roberts A.M.
      • Miyamoto D.K.
      • Ku W.M.
      • Huffman T.R.
      • Petri Y.
      • Heslin M.J.
      • Contreras C.M.
      • Skibola C.F.
      • Olzmann J.A.
      • Nomura D.K.
      Chemoproteomics-enabled covalent ligand screen reveals a cysteine hotspot in reticulon 4 that impairs ER morphology and cancer pathogenicity.
      ). Reduced labeling in cell lysates, as observed clearly in our data, may be caused by conformational changes resulting from ER membrane disruption or oxidative modification of the Cys1101 upon cell lysis.

      Substantial differences in lysate and intact cell labeling relate to redox-sensitive cysteines on proteins involved in redox signaling

      We observed that several functional cysteines from proteins involved in redox signaling were labeled by the ABP very distinctively in the cell lysates compared to intact cells (Supplementary Data 1) (Figure 3C). Among these proteins were found to be several members of the peroxiredoxin (PRDX) family (PRDX1, PRDX3, PRDX4, PRDX6), thioredoxin (TXN), glutaredoxin (GLRX) and glutathione S-transferase P (GSTP1) (
      • Poole L.B.
      The basics of thiols and cysteines in redox biology and chemistry.
      ). Three PRDX proteins, PRDX1, PRDX3, and PRDX4, detected are 2-Cys PRDXs that reduce H2O2 by donating an electron from the peroxidatic cysteine (
      • Elko E.A.
      • Cunniff B.
      • Seward D.J.
      • Chia S.B.
      • Aboushousha R.
      • Van De Wetering C.
      • Van Der Velden J.
      • Manuel A.
      • Shukla A.
      • Heintz N.H.
      • Anathy V.
      • Van Der Vliet A.
      • Janssen-Heininger Y.M.W.
      Peroxiredoxins and Beyond; Redox Systems Regulating Lung Physiology and Disease.
      ). The carboxy-terminal cysteine from another 2-cys peroxiredoxin can consecutively form a disulfide bond to generate a PRDX-dimer. In PRDX1, PRDX3 and PRDX4, the conserved carboxy-terminal resolving cysteines are Cys173, Cys229 and Cys245, respectively. TXN is involved in reducing this disulfide bond (
      • Elko E.A.
      • Cunniff B.
      • Seward D.J.
      • Chia S.B.
      • Aboushousha R.
      • Van De Wetering C.
      • Van Der Velden J.
      • Manuel A.
      • Shukla A.
      • Heintz N.H.
      • Anathy V.
      • Van Der Vliet A.
      • Janssen-Heininger Y.M.W.
      Peroxiredoxins and Beyond; Redox Systems Regulating Lung Physiology and Disease.
      ). Whether this carboxy-terminal resolving cysteine of PRDX is free or involved in a disulfide bridge to form a dimer depends on various factors, such as concentrations of H2O2 and TXN (
      • Elko E.A.
      • Cunniff B.
      • Seward D.J.
      • Chia S.B.
      • Aboushousha R.
      • Van De Wetering C.
      • Van Der Velden J.
      • Manuel A.
      • Shukla A.
      • Heintz N.H.
      • Anathy V.
      • Van Der Vliet A.
      • Janssen-Heininger Y.M.W.
      Peroxiredoxins and Beyond; Redox Systems Regulating Lung Physiology and Disease.
      ). Consistently in all our ABP labeling replicates the carboxy-terminal cysteines of PRDX1, PRDX3 and PRDX4 are found to be intensely labeled in intact cells, and either not detected or low-level labeled in the cell lysates, indicating that these free cysteines become oxidized and substantially less available for ABP labeling in the cell lysates (Figure 3C).
      While we observe similar probe-labeling intensities for peroxidatic Cys47 in 1-cys peroxiredoxin 6 in intact cells, a loss or decrease of signal of this site in lysates is not observed. On the contrary, Cys47 was more abundantly labeled in lysate compared to intact cells. Cys48 on GSTP1, a partner for heterodimerization with PRDX6, which acts as the resolving cysteine for PRDX6, shows a similar trend (
      • Ralat L.A.
      • Misquitta S.A.
      • Manevich Y.
      • Fisher A.B.
      • Colman R.F.
      Characterization of the complex of glutathione S-transferase pi and 1-cysteine peroxiredoxin.
      ,
      • Fisher A.B.
      Peroxiredoxin 6: A Bifunctional Enzyme with Glutathione Peroxidase and Phospholipase A2 Activities.
      ). GSTP1 heterodimerizes with oxidized PRDX6 and catalyzes the S-glutathionylation of Cys47, during which a disulfide bridge between Cys47 on PRDX6 and Cys48 on GSTP1 can occur (
      • Ralat L.A.
      • Misquitta S.A.
      • Manevich Y.
      • Fisher A.B.
      • Colman R.F.
      Characterization of the complex of glutathione S-transferase pi and 1-cysteine peroxiredoxin.
      ). To acquire a reduced PRDX6, PRDX6-GSH (glutathione) interacts with another GSH molecule to form oxidized GSH (
      • Elko E.A.
      • Cunniff B.
      • Seward D.J.
      • Chia S.B.
      • Aboushousha R.
      • Van De Wetering C.
      • Van Der Velden J.
      • Manuel A.
      • Shukla A.
      • Heintz N.H.
      • Anathy V.
      • Van Der Vliet A.
      • Janssen-Heininger Y.M.W.
      Peroxiredoxins and Beyond; Redox Systems Regulating Lung Physiology and Disease.
      ,
      • Zhou S.
      • Lien Y.C.
      • Shuvaeva T.
      • Debolt K.
      • Feinstein S.I.
      • Fisher A.B.
      Functional interaction of glutathione S-transferase pi and peroxiredoxin 6 in intact cells.
      ). More intense labeling in lysate by the ABP, might indicate that PRDX6 and GSTP1 are not oxidized upon lysis, unlike the 2-cys peroxiredoxins and could thus be probe targeted.
      Thioredoxin (TXN), is abundantly labeled on Cys73, again only in intact cells. Cys73 is involved in the formation of TXN homo-dimers and thereby regulates TXN activity (
      • Hashemy S.I.
      • Holmgren A.
      Regulation of the Catalytic Activity and Structure of Human Thioredoxin 1 via Oxidation and S-Nitrosylation of Cysteine Residues.
      ). In addition to dimer formation, the activity of TXN can also be regulated by S-nitrosylation or S-glutathiolation of Cys73 (
      • Han S.
      Molecular dynamics simulations of thioredoxin with S-glutathiolated cysteine-73.
      ,
      • Casagrande S.
      • Bonetto V.
      • Fratelli M.
      • Gianazza E.
      • Eberini I.
      • Massignan T.
      • Salmona M.
      • Chang G.
      • Holmgren A.
      • Ghezzi P.
      • Herzenberg L.A.
      Glutathionylation of human thioredoxin: A possible crosstalk between the glutathione and thioredoxin systems.
      ). Moreover, PX-12, a TXN inhibitor, is also proposed to exert its effect through binding at Cys73 (
      • Mao Z.
      • Yang X.
      • Mizutani S.
      • Huang Y.
      • Zhang Z.
      • Shinmori H.
      • Gao K.
      • Yao J.
      Hydrogen Sulfide Mediates Tumor Cell Resistance to Thioredoxin Inhibitor.
      ). The abundant ABP binding of TXN Cys73 in intact cells, and the lack of labeling in cell lysates could again be due to the free cysteine being oxidized or modified differently upon formation of the lysate (Figure 3C).
      Finally, in glutaredoxin (GLRX) we observe Cys8 to be abundantly ABP labeled in intact cells in all intact cell replicates, but not in cell lysates. GLRX catalyzes deglutathionylation through the nucleophilic displacement of the GSH moiety by the active site cysteine, followed by the rate-limiting step where the thiolate ion of the active site cysteine is regenerated consuming one molecule of GSH (
      • Chia S.B.
      • Elko E.A.
      • Aboushousha R.
      • Manuel A.M.
      • van de Wetering C.
      • Druso J.E.
      • van der Velden J.
      • Seward D.J.
      • Anathy V.
      • Irvin C.G.
      • Lam Y.W.
      • van der Vliet A.
      • Janssen-Heininger Y.M.W.
      Dysregulation of the glutaredoxin/S-glutathionylation redox axis in lung diseases.
      ). Cys8 has been linked to oxidative inactivation of GLRX and has also been reported to potentially be modified via S-nitrosylation (
      • Hashemy S.I.
      • Johansson C.
      • Berndt C.
      • Lillig C.H.
      • Holmgren A.
      Oxidation and S-Nitrosylation of Cysteines in Human Cytosolic and Mitochondrial Glutaredoxins: EFFECTS ON STRUCTURE AND ACTIVITY.
      ). The abundant ABP binding of GLRX Cys8 in intact cells, and the lack of labeling in cell lysates could again be due to the free Cys8 becoming rapidly modified upon formation of the lysate (Figure 3C).

      PF-06672131 is directed towards nucleotide binding pockets

      Afatinib and its derived ABP PF-06672131 have been shown to bind in an ATP binding pocket of EGFR (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ,
      • Solca F.
      • Dahl G.
      • Zoephel A.
      • Bader G.
      • Sanderson M.
      • Klein C.
      • Kraemer O.
      • Himmelsbach F.
      • Haaksma E.
      • Adolf G.R.
      Target binding properties and cellular activity of afatinib (BIBW 2992), an irreversible ErbB family blocker.
      ). Therefore it has been hypothesized that PF-06672131 mimics ATP, and may also display affinity for ATP binding pockets belonging to proteins other than EGFR (
      • Lanning B.R.
      • Whitby L.R.
      • Dix M.M.
      • Douhan J.
      • Gilbert A.M.
      • Hett E.C.
      • Johnson T.O.
      • Joslyn C.
      • Kath J.C.
      • Niessen S.
      • Roberts L.R.
      • Schnute M.E.
      • Wang C.
      • Hulce J.J.
      • Wei B.
      • Whiteley L.O.
      • Hayward M.M.
      • Cravatt B.F.
      A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
      ,
      • Shindo N.
      • Fuchida H.
      • Sato M.
      • Watari K.
      • Shibata T.
      • Kuwata K.
      • Miura C.
      • Okamoto K.
      • Hatsuyama Y.
      • Tokunaga K.
      • Sakamoto S.
      • Morimoto S.
      • Abe Y.
      • Shiroishi M.
      • Caaveiro J.M.M.
      • Ueda T.
      • Tamura T.
      • Matsunaga N.
      • Nakao T.
      • Koyanagi S.
      • Ohdo S.
      • Yamaguchi Y.
      • Hamachi I.
      • Ono M.
      • Ojida A.
      Selective and reversible modification of kinase cysteines with chlorofluoroacetamides.
      ). We indeed observed that around 20% of the detected probe-targeted proteins are ATP binders, which corresponds to a 2.5-fold enrichment compared to the percentage of ATP binders in the total human proteome as determined by GO term molecular function overrepresentation analysis (PANTHER, Protein analysis through evolutionary relationships) (
      • Mi H.
      • Muruganujan A.
      • Huang X.
      • Ebert D.
      • Mills C.
      • Guo X.
      • Thomas P.D.
      Protocol Update for large-scale genome and gene function analysis with the PANTHER classification system (v.14.0).
      ,
      • Mi H.
      • Ebert D.
      • Muruganujan A.
      • Mills C.
      • Albou L.P.
      • Mushayamaha T.
      • Thomas P.D.
      PANTHER version 16: a revised family classification, tree-based classification tool, enhancer regions and extensive API.
      ). In addition to ATP binders, around 10% of the probe targets are known to bind other nucleotides, including GTP (6%) and NAD (3%) (Figure 4A). Together these observations point to a preferential targeting of PF-06672131 towards ATP and other nucleotide binding proteins.
      Figure thumbnail gr4
      Figure 4PF-06672131 binds nucleotide binding proteins, including the Ras superfamily. A. Bar graph of the number of different nucleotide (ATP, GTP, NAD(P), FAD) binding proteins detected to be probe-bound in cell lysates and intact cells, about 30% of the ABP-bound proteins could be classified as nucleotide binding proteins (PANTHER database (
      • Mi H.
      • Ebert D.
      • Muruganujan A.
      • Mills C.
      • Albou L.P.
      • Mushayamaha T.
      • Thomas P.D.
      PANTHER version 16: a revised family classification, tree-based classification tool, enhancer regions and extensive API.
      ,
      • Thomas P.D.
      • Campbell M.J.
      • Kejariwal A.
      • Mi H.
      • Karlak B.
      • Daverman R.
      • Diemer K.
      • Muruganujan A.
      • Narechania A.
      PANTHER: a library of protein families and subfamilies indexed by function.
      )). B. Cysteine residues found to be reacting with PF-06672131 of proteins belonging to the Ras superfamily. C. Docking of PF-06672131 onto Cys16 in the GTP-binding site in a crystal structure of RHOA (PDB: 1FTN) (
      • Wei Y.
      • Zhang Y.
      • Derewenda U.
      • Liu X.
      • Minor W.
      • Nakamoto R.K.
      • Somlyo A.V.
      • Somlyo A.P.
      • Derewenda Z.S.
      Crystal structure of RhoA–GDP and its functional implications.
      ). PF-06672131 is indicated in green and GTP is indicated in red. D. Protein ligand interaction profiler (PLIP) web server analysis reveals that next to the covalent bond to cysteine (green line), PF-06672131 (orange) might interact with RHOA (grey) through 4 hydrogen bonds (blue lines), 7 hydrophobic interactions (red dashed lines) and 1 salt bridge (yellow dashed line) () (
      • Adasme M.F.
      • Linnemann K.L.
      • Bolz S.N.
      • Kaiser F.
      • Salentin S.
      • Haupt V.J.
      • Schroeder M.
      PLIP 2021: expanding the scope of the protein–ligand interaction profiler to DNA and RNA.
      ).

      PF-06672131 binds to conserved cysteines in GTP binding pockets from the Ras superfamily of GTPases

      Ras super family proteins are involved in targeting and regulation of vesicular membrane trafficking (
      • Wennerberg K.
      • Rossman K.L.
      • Der C.J.
      The Ras superfamily at a glance.
      ,
      • Mukhopadhyay S.
      • Vander Heiden M.G.
      • McCormick F.
      The metabolic landscape of RAS-driven cancers from biology to therapy.
      ). These small GTPase proteins act as molecular switches that are turned on by guanosine exchange factors (GEFs) that catalyze the conversion of the GDP to the GTP bound state and are ‘switched off’ by GTPase-activating proteins (GAPs), which enhance GTP hydrolysis to GDP. In their active GTP bound state, RAS super family proteins recruit effector proteins through which they exert their biological effects (
      • Wennerberg K.
      • Rossman K.L.
      • Der C.J.
      The Ras superfamily at a glance.
      ,
      • Simanshu D.K.
      • Nissley D.V.
      • McCormick F.
      RAS Proteins and Their Regulators in Human Disease.
      ). Since the employed ABP is specifically directed towards cysteine residues in proteins, it is important to note that besides regulation by GAPs and GEFs, small GTPases are also highly and tightly regulated by various cysteine post-translational modifications, including farnesylation, S-palmitoylation, glutathionylation, and disulfide bridge formation (
      • Aaron Hobbs G.
      • Zhou B.
      • Cox A.D.
      • Campbell S.L.
      Rho GTPases, oxidation, and cell redox control.
      ,
      • Cherfils J.
      • Zeghouf M.
      Regulation of Small GTPases by GEFs, GAPs, and GDIs.
      ,
      • Hodge R.G.
      • Ridley A.J.
      Regulating Rho GTPases and their regulators.
      ,
      • Hurst M.
      • McGarry D.J.
      • Olson M.F.
      Rho GTPases: Non-canonical regulation by cysteine oxidation.
      ,
      • Pylypenko O.
      • Hammich H.
      • Yu I.M.
      • Houdusse A.
      Rab GTPases and their interacting protein partners: Structural insights into Rab functional diversity.
      ).
      The nuclear small GTPase Ran was found to contain two ABP labeling sites within the same tryptic peptide, on Cys112 and Cys120 (Figure 4B, Supplementary Data 1). Nevertheless, localization and intensity of probe-binding could be distinguished and it was revealed that Cys120 was found to be more intensely labeled in lysate and in intact cells (Supplementary Data 1). A previous study hypothesized that Cys112 is an oxidation target by pervanadate and causes degradation of RAN (
      • Tao G.Z.
      • Zhou Q.
      • Strnad P.
      • Salemi M.R.
      • Young M.L.
      • Omary M.B.
      Human Ran Cysteine 112 Oxidation by Pervanadate Regulates Its Binding to Keratins.
      ). Cys120, on the other hand, was shown not to be a target of pervanadate-mediated oxidation. Cys120 being more labeled in both conditions might indicate that Cys112 is more frequently oxidized and therefore, less available for ABP binding in both lysate and intact cells.
      Within the Rab small GTPase protein family, the ABP targets a conserved cysteine in the nucleotide binding site of RAB1A (Cys126), RAB8A (Cys123) and RAB13 (Cys123), which are all labeled substantially more in cell lysates compared to intact cells, consistent with the hypothesis of reduced competition with endogenous nucleotides due to lower local nucleotide concentrations in lysates compared to intact cells (
      • Savitski M.M.
      • Reinhard F.B.M.
      • Franken H.
      • Werner T.
      • Savitski M.F.
      • Eberhard D.
      • Molina D.M.
      • Jafari R.
      • Dovega R.B.
      • Klaeger S.
      • Kuster B.
      • Nordlund P.
      • Bantscheff M.
      • Drewes G.
      Tracking cancer drugs in living cells by thermal profiling of the proteome.
      ). The observed probe-labeled cysteines in RAB3A/D (Cys184) and RAB3B (Cys184) are not situated in the GTP binding pocket, but are surface exposed and known to be next to the complementary determining region 3 which plays a role in the binding of the small GTPase to its effectors. (
      • Pylypenko O.
      • Hammich H.
      • Yu I.M.
      • Houdusse A.
      Rab GTPases and their interacting protein partners: Structural insights into Rab functional diversity.
      ) In a co-crystal structure of Rattus Norvegicus Rab3A bound to its effector protein Rabphilin (PDB:1ZBD), Cys184 is positioned at the binding interface of the two proteins (
      • Ostermeier C.
      • Brunger A.T.
      Structural Basis of Rab Effector Specificity: Crystal Structure of the Small G Protein Rab3A Complexed with the Effector Domain of Rabphilin-3A.
      ). Therefore binding of Cys184 by the ABP could potentially also block the interaction between RAB3A and Rabphilin.
      The Rho GTPase family forms part of the Ras superfamily that regulate a wide range of cellular responses, including cell adhesion and changes to the cytoskeleton (
      • Hodge R.G.
      • Ridley A.J.
      Regulating Rho GTPases and their regulators.
      ). Multiple members of this subfamily are also labeled at the conserved cysteine residues Cys159 in RHOA, RHOC, and Cys157 in CDC42, within their nucleotide binding pockets. RAC1, RAC2 and RAC3 appear to also be targeted at Cys157. However, the relative labeling efficiency between RAC1, 2 and 3, between the isoforms cannot be determined as they share identical Cys157 containing tryptic peptides. These cysteines all show preferential labeling in the cell lysate, which again, is in line with lower concentrations of competing nucleotides in cell lysates. Cys157 in RAC1 is known to be regulated by glutathionylation during metabolic stress, which is proposed to have an inhibitory effect on RAC1 activity (
      • Han J.
      • Weisbrod R.M.
      • Shao D.
      • Watanabe Y.
      • Yin X.
      • Bachschmid M.M.
      • Seta F.
      • Janssen-Heininger Y.M.W.
      • Matsui R.
      • Zang M.
      • Hamburg N.M.
      • Cohen R.A.
      The redox mechanism for vascular barrier dysfunction associated with metabolic disorders: Glutathionylation of Rac1 in endothelial cells.
      ).
      Moreover, other cysteines in the Rho GTPase family that undergo posttranslational modifications are targeted by the ABP. For example Cys16 of RHOA, which is in close proximity of Cys157 in the GTP binding pocket, can inactivate the protein by formation of a disulfide bond with another cysteine (Cys20) (
      • Hurst M.
      • McGarry D.J.
      • Olson M.F.
      Rho GTPases: Non-canonical regulation by cysteine oxidation.
      ). Probe labeling of this specific Cys16 was identified in a tryptic peptide that is shared by RHOA and RHOC in both cell lysates and intact cells. In addition, we identified labeling of Cys178 on a tryptic peptide shared by RAC1 and RAC3, this cysteine is a known target for S-palmitoylation which regulates protein localization and affects GTP binding. (
      • Navarro-Lérida I.
      • Sánchez-Perales S.
      • Calvo M.
      • Rentero C.
      • Zheng Y.
      • Enrich C.
      • Del Pozo M.A.
      A palmitoylation switch mechanism regulates Rac1 function and membrane organization.
      ).
      Together these probe interactions reveal that PF-06672131 targets multiple sites in the Ras superfamily including sites residing in nucleotide binding pockets and sites that undergo posttranslational modifications. Knowledge of the exact ABP binding site can guide predictions of the functional effect of inhibitor treatment for specific proteins. In addition, the observations in the Rho family uncover the limitation that the exact identity of a protein target cannot always be unequivocally determined based on single peptides, especially for proteins with close homologues. This limitation might however be resolved by using other proteases in addition to trypsin to generate unique peptides.

      In silico binding pose prediction in RHOA

      To simulate the interaction between PF-06672131 and RHOA, HADDOCK 2.4 was used to dock the ABP on Cys16 in the GDP/GTP binding pocket on a RHOA crystal structure (PDB: 1FTN). (
      • Honorato R.V.
      • Koukos P.I.
      • Jiménez-García B.
      • Tsaregorodtsev A.
      • Verlato M.
      • Giachetti A.
      • Rosato A.
      • Bonvin A.M.J.J.
      Structural Biology in the Clouds: The WeNMR-EOSC Ecosystem.
      ,
      • Wei Y.
      • Zhang Y.
      • Derewenda U.
      • Liu X.
      • Minor W.
      • Nakamoto R.K.
      • Somlyo A.V.
      • Somlyo A.P.
      • Derewenda Z.S.
      Crystal structure of RhoA–GDP and its functional implications.
      ,
      • Van Zundert G.C.P.
      • Rodrigues J.P.G.L.M.
      • Trellet M.
      • Schmitz C.
      • Kastritis P.L.
      • Karaca E.
      • Melquiond A.S.J.
      • Van Dijk M.
      • De Vries S.J.
      • Bonvin A.M.J.J.
      The HADDOCK2.2 Web Server: User-Friendly Integrative Modeling of Biomolecular Complexes.
      ). The best-fitting docking pose with the lowest distance of the carbon of PF-06672131 to the Cys16 sulfur atom shows that the ABP fits well in the RHOA nucleotide binding pocket (Figure 4C). A distance of 3.0 Ångström for the sulfur-carbon bond between the ABP and Cys16 of the protein was measured. Given the resolution of 2.2 Ångström, and a C – S bond being 1.8 Ångström, this falls within the error margin. Additional analysis of the interactions between PF-06672131 and RHOA shows that the probe might bind in the GTP binding pocket through four hydrogen bonds, seven hydrophobic interactions and a salt bridge, hinting that affinity of the probe for the GTP binding pocket may partially originate from these interactions (Figure 4D, Supplementary Data 1) (
      • Adasme M.F.
      • Linnemann K.L.
      • Bolz S.N.
      • Kaiser F.
      • Salentin S.
      • Haupt V.J.
      • Schroeder M.
      PLIP 2021: expanding the scope of the protein–ligand interaction profiler to DNA and RNA.
      ). The in silico binding pose of the ABP in RHOA indicates that computational efforts guided by our mass spectrometry data may gain insight of inhibitor binding poses. Knowledge on the exact binding site of the ABP significantly limits the possible binding poses, increasing the accuracy of the model. Therefore, ABPP-PhosID in combination with computational modelling can contribute to drug development by guiding inhibitor optimization.

      Pepsin digestion allows detection of PF-06672131 binding to non-catalytic active site cysteine in EGFR

      Initially, we did not detect binding of PF-06672131 to its known target, Cys797 in the ATP binding pocket of EGFR. We hypothesized that the size and the hydrophobic nature of the generated tryptic peptide might hamper detection using our proteomics LC-MS/MS methodology (27 amino acids, Grand average of hydropathy, GRAVY: 1.19, Figure 5B). Therefore, we did explore other proteases for the digestion, and found especially pepsin very useful. Using pepsin as protease, probe labeled Cys797 containing peptides were abundantly detected in intact cells with 161 PSMs, originating from 12 different peptides (Figure 5B, Supplementary Data 4). This high amount of different pepsin-derived peptides can be explained by the more diverse cleavage specificity of pepsin compared to trypsin. Comparing the target landscape between pepsin- and trypsin-mediated PhosID–ABPP revealed that the observed overlap in probe-bound cysteines is relatively low (189 sites) and the pepsin-based approach revealed an extra 675 ABP binding sites (Figure 5A). Thus, strategies using alternative proteases can strengthen PhosID–ABPP by expanding the landscape of detected binding proteins and their binding sites. Moreover, the evidence for specific probe binding sites can be improved using two or more proteases to generate multiple peptides containing the site of ABP-protein interaction. For example Cys1101 on RTN4 and Cys173 on Peroxiredoxin-1 (Figure 5B). Differential enzymatic digestion also provides the opportunity to distinguish probe-bound cysteines between variants of proteins with high homology. For example, PRDX1 and -4 share high homologous regions around the active site cysteines Cys173 and Cys245 respectively, and could not be discerned with the tryptic peptide (HGEVCPAGWK, Figure 5B). Using pepsin, these variants could be distinguished with peptides that are unique for peroxiredoxin-1 or -4, indicating that Cys173 on PRDX1 was more intensely labeled than Cys245 on PRDX4, as the total number of PSMs for Cys173 on PRDX1 were found to be higher (Figure 5B). Together, these data show that alternative protease strategies strengthen PhosID–ABPP by revealing novel binding sites, improving the evidence for ABP binding sites through different enzymatic cleavages, and distinguish ABP binding sites on different proteins sharing high homology.
      Figure thumbnail gr5
      Figure 5proteolysis with pepsin widens the landscape of identified ABP-binding sites, including the site on EGFR. A. Venn diagram indicating the overlap of ABP binding sites found by trypsin (grey) and pepsin proteolysis (green) of probe treated intact cells (Supplementary Data 4). Pepsin expands the scope of PhosID-ABPP, with 675 sites exclusively detected with the pepsin-based approach, including the supposed primary binding target and site of PF-06672131 on EGFR Cys797. 189 probe-targeted cysteines were found in common when using trypsin and pepsin. B. Number of psms found for evidence of individual ABP-bound peptides. Peptides with ABP binding site Cys797 on EGFR, Cys1101 on RTN4, Cys173 on PRDX1, and Cys245 on PRDX4 through a trypsin (grey) and pepsin (green) approach are shown. The red cysteines indicate the detected binding site of PF-06672131. The orange annotated amino acid letters indicate difference in sequence between PRDX1 and PRDX4.

      Conclusion

      Here we introduce a new ABPP approach based on employing IMAC-enrichable phosphonate affinity tags, allowing the efficient and selective enrichment of peptides bound to an activity-based probe (ABP), with as clear benefit enabling the identification of the exact drug binding site. Using PhosID–ABPP, we were able to robustly and reproducibly detect hundreds of protein targets, charting the site-specific target landscape of the alkynylated Afatinib derivative ABP (PF-06672131) in human A431 skin cancer cells. The exact binding site profile revealed that the ABP not only reacted with active site cysteines in the ATP binding pocket of EGFR, but also to many surface accessible cysteines and cysteines in binding pockets of not only ATP but also other nucleotides. PhosID–ABPP was applied to both intact cells and cellular lysates. Many differences in labeling of specific cysteines were observed between intact cells and cell lysates. Some of these differences may originate from redox sensitive cysteines undergoing oxidation-reduction cycles during protein function and differences in nucleotide binding status of proteins in lysates versus intact cells. Through the use of pepsin as an alternative protease, instead of the commonly used trypsin, we detected the main EGFR target binding site of PF-06672131 with multiple different peptides, demonstrating that the utility of PhosID-ABPP could be further augmented by orthogonal use of proteolytic enzymes.
      PhosID-ABPP allows future studies on site specific drug occupancy by concentration and time dependent competition with the parent inhibitor Afatinib, and we anticipate that the workflow can be readily extended to other ABPs. With this, we envision PhosID-ABPP as a highly valuable addition to the ABPP toolbox, which is also complementary to existing protein-level enrichment strategies.

      Data availability

      The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD036569. This article contains supplemental data: Supplementary Figures 1-3, Supplementary Table 1, and Supplementary Data Excel files 1-4.

      Acknowledgments

      We acknowledge support from the Netherlands Organization for Scientific Research (NWO) funding the Netherlands Proteomics Centre through the X-omics Road Map program (project 184.034.019).

      References

        • Cravatt B.F.
        • Wright A.T.
        • Kozarich J.W.
        Activity-Based Protein Profiling: From Enzyme Chemistry to Proteomic Chemistry.
        Annu. Rev. Biochem. 2008; 77: 383-414
        • Parker C.G.
        • Pratt M.R.
        Click Chemistry in Proteomic Investigations.
        Cell. 2020; 180: 605-632
        • Willems L.I.
        • Overkleeft H.S.
        • Van Kasteren S.I.
        Current developments in activity-based protein profiling.
        Bioconjug. Chem. 2014; 25: 1181-1191
        • Serwa R.
        • Tate E.W.
        Activity-Based Profiling for Drug Discovery.
        Chem. Biol. 2011; 18: 407-409
        • Benns H.J.
        • Wincott C.J.
        • Tate E.W.
        • Child M.A.
        Activity- and reactivity-based proteomics: Recent technological advances and applications in drug discovery.
        Curr. Opin. Chem. Biol. 2021; 60: 20-29
        • Van Esbroeck A.C.M.
        • Janssen A.P.A.
        • Cognetta A.B.
        • Ogasawara D.
        • Shpak G.
        • Van Der Kroeg M.
        • Kantae V.
        • Baggelaar M.P.
        • De Vrij F.M.S.
        • Deng H.
        • Allarà M.
        • Fezza F.
        • Lin Z.
        • Van Der Wel T.
        • Soethoudt M.
        • Mock E.D.
        • Den Dulk H.
        • Baak I.L.
        • Florea B.I.
        • Hendriks G.
        • De Petrocellis L.
        • Overkleeft H.S.
        • Hankemeier T.
        • De Zeeuw C.I.
        • Di Marzo V.
        • Maccarrone M.
        • Cravatt B.F.
        • Kushner S.A.
        • Van Der Stelt M.
        Activity-based protein profiling reveals off-target proteins of the FAAH inhibitor BIA 10-2474.
        Science (80-. ). 2017; 356: 1084-1087
        • Liu Y.
        • Patricelli M.P.
        • Cravatt B.F.
        Activity-based protein profiling: The serine hydrolases.
        Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 14694-14699
        • Greenbaum D.
        • Baruch A.
        • Hayrapetian L.
        • Darula Z.
        • Burlingame A.
        • Medzihradszky K.F.
        • Bogyo M.
        Chemical Approaches for Functionally Probing the Proteome.
        Mol. Cell. Proteomics. 2002; 1: 60-68
        • Meissner F.
        • Geddes-McAlister J.
        • Mann M.
        • Bantscheff M.
        The emerging role of mass spectrometry-based proteomics in drug discovery.
        Nat. Rev. Drug Discov. 2022; https://doi.org/10.1038/s41573-022-00409-3
        • Altelaar A.F.M.
        • Munoz J.
        • Heck A.J.R.
        Next-generation proteomics: towards an integrative view of proteome dynamics.
        Nat. Rev. Genet. 2013; 14: 35-48
        • Schastnaya E.
        • Raguz Nakic Z.
        • Gruber C.H.
        • Doubleday P.F.
        • Krishnan A.
        • Johns N.I.
        • Park J.
        • Wang H.H.
        • Sauer U.
        (2021) Extensive regulation of enzyme activity by phosphorylation in Escherichia coli.
        Nat. Commun. 2021; 12: 5650
        • Holdgate G.A.
        • Meek T.D.
        • Grimley R.L.
        Mechanistic enzymology in drug discovery: a fresh perspective.
        Nat. Rev. Drug Discov. 2017; 17: 115-132
        • Motlagh H.N.
        • Wrabl J.O.
        • Li J.
        • Hilser V.J.
        The ensemble nature of allostery.
        Nature. 2014; 508: 331-339
        • Lanning B.R.
        • Whitby L.R.
        • Dix M.M.
        • Douhan J.
        • Gilbert A.M.
        • Hett E.C.
        • Johnson T.O.
        • Joslyn C.
        • Kath J.C.
        • Niessen S.
        • Roberts L.R.
        • Schnute M.E.
        • Wang C.
        • Hulce J.J.
        • Wei B.
        • Whiteley L.O.
        • Hayward M.M.
        • Cravatt B.F.
        A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors.
        Nat. Chem. Biol. 2014; 10: 760-767
        • Fang H.
        • Peng B.
        • Ong S.Y.
        • Wu Q.
        • Li L.
        • Yao S.Q.
        Recent advances in activity-based probes (ABPs) and affinity-based probes (AfBPs) for profiling of enzymes.
        Chem. Sci. 2021; 12: 8288-8310
        • Bird R.E.
        • Lemmel S.A.
        • Yu X.
        • Zhou Q.A.
        Bioorthogonal Chemistry and Its Applications.
        Bioconjug. Chem. 2021; 32: 2457-2479
        • Schiapparelli L.M.
        • McClatchy D.B.
        • Liu H.-H.
        • Sharma P.
        • Yates J.R.
        • Cline H.T.
        Direct Detection of Biotinylated Proteins by Mass Spectrometry.
        J. Proteome Res. 2014; 13: 3966-3978
        • Browne C.M.
        • Jiang B.
        • Ficarro S.B.
        • Doctor Z.M.
        • Johnson J.L.
        • Card J.D.
        • Sivakumaren S.C.
        • Alexander W.M.
        • Yaron T.M.
        • Murphy C.J.
        • Kwiatkowski N.P.
        • Zhang T.
        • Cantley L.C.
        • Gray N.S.
        • Marto J.A.
        A Chemoproteomic Strategy for Direct and Proteome-Wide Covalent Inhibitor Target-Site Identification.
        J. Am. Chem. Soc. 2019; 141: 191-203
        • Zanon P.R.A.
        • Lewald L.
        • Hacker S.M.
        Isotopically Labeled Desthiobiotin Azide (isoDTB) Tags Enable Global Profiling of the Bacterial Cysteinome.
        Angew. Chemie Int. Ed. 2020; 59: 2829-2836
        • Weerapana E.
        • Wang C.
        • Simon G.M.
        • Richter F.
        • Khare S.
        • Dillon M.B.D.
        • Bachovchin D.A.
        • Mowen K.
        • Baker D.
        • Cravatt B.F.
        Quantitative reactivity profiling predicts functional cysteines in proteomes.
        Nature. 2010; 468: 790-797
        • Weerapana E.
        • Speers A.E.
        • Cravatt B.F.
        Tandem orthogonal proteolysis-activity-based protein profiling (TOP-ABPP)—a general method for mapping sites of probe modification in proteomes.
        Nat. Protoc. 2007; 2: 1414-1425
        • Post H.
        • Penning R.
        • Fitzpatrick M.A.
        • Garrigues L.B.
        • Wu W.
        • Macgillavry H.D.
        • Hoogenraad C.C.
        • Heck A.J.R.
        • Altelaar A.F.M.
        Robust, Sensitive, and Automated Phosphopeptide Enrichment Optimized for Low Sample Amounts Applied to Primary Hippocampal Neurons.
        J. Proteome Res. 2017; 16: 728-737
        • Kleinpenning F.
        • Steigenberger B.
        • Wu W.
        • Heck A.J.R.
        (2020) Fishing for newly synthesized proteins with phosphonate-handles.
        Nat. Commun. 2020; 11: 3244
        • Shindo N.
        • Fuchida H.
        • Sato M.
        • Watari K.
        • Shibata T.
        • Kuwata K.
        • Miura C.
        • Okamoto K.
        • Hatsuyama Y.
        • Tokunaga K.
        • Sakamoto S.
        • Morimoto S.
        • Abe Y.
        • Shiroishi M.
        • Caaveiro J.M.M.
        • Ueda T.
        • Tamura T.
        • Matsunaga N.
        • Nakao T.
        • Koyanagi S.
        • Ohdo S.
        • Yamaguchi Y.
        • Hamachi I.
        • Ono M.
        • Ojida A.
        Selective and reversible modification of kinase cysteines with chlorofluoroacetamides.
        Nat. Chem. Biol. 2019; 15: 250-258
        • Cox J.
        • Mann M.
        MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.
        Nat. Biotechnol. 2008; 26: 1367-1372
        • Kong A.T.
        • Leprevost F.V.
        • Avtonomov D.M.
        • Mellacheruvu D.
        • Nesvizhskii A.I.
        MSFragger: Ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics.
        Nat. Methods. 2017; 14: 513-520
        • Tyanova S.
        • Temu T.
        • Sinitcyn P.
        • Carlson A.
        • Hein M.Y.
        • Geiger T.
        • Mann M.
        • Cox J.
        The Perseus computational platform for comprehensive analysis of (prote)omics data.
        Nat. Methods. 2016; 13: 731-740
        • Hulsen T.
        • de Vlieg J.
        • Alkema W.
        BioVenn - A web application for the comparison and visualization of biological lists using area-proportional Venn diagrams.
        BMC Genomics. 2008; 9: 1-6
        • Solca F.
        • Dahl G.
        • Zoephel A.
        • Bader G.
        • Sanderson M.
        • Klein C.
        • Kraemer O.
        • Himmelsbach F.
        • Haaksma E.
        • Adolf G.R.
        Target binding properties and cellular activity of afatinib (BIBW 2992), an irreversible ErbB family blocker.
        J. Pharmacol. Exp. Ther. 2012; 343: 342-350
        • Rodrigues J.P.G.L.M.
        • Teixeira J.M.C.
        • Trellet M.
        • Bonvin A.M.J.J.
        pdb-tools: a swiss army knife for molecular structures.
        F1000Research. 2018; 7: 1961
        • Jiménez-García B.
        • Teixeira J.M.C.
        • Trellet M.
        • Rodrigues J.P.G.L.M.
        • Bonvin A.M.J.J.
        PDB-tools web: A user-friendly interface for the manipulation of PDB files.
        Proteins Struct. Funct. Bioinforma. 2021; 89: 330-335
        • Honorato R.V.
        • Koukos P.I.
        • Jiménez-García B.
        • Tsaregorodtsev A.
        • Verlato M.
        • Giachetti A.
        • Rosato A.
        • Bonvin A.M.J.J.
        Structural Biology in the Clouds: The WeNMR-EOSC Ecosystem.
        Front. Mol. Biosci. 2021; 8: 708
        • Wei Y.
        • Zhang Y.
        • Derewenda U.
        • Liu X.
        • Minor W.
        • Nakamoto R.K.
        • Somlyo A.V.
        • Somlyo A.P.
        • Derewenda Z.S.
        Crystal structure of RhoA–GDP and its functional implications.
        Nat. Struct. Biol. 1997; 4: 699-703
        • Van Zundert G.C.P.
        • Rodrigues J.P.G.L.M.
        • Trellet M.
        • Schmitz C.
        • Kastritis P.L.
        • Karaca E.
        • Melquiond A.S.J.
        • Van Dijk M.
        • De Vries S.J.
        • Bonvin A.M.J.J.
        The HADDOCK2.2 Web Server: User-Friendly Integrative Modeling of Biomolecular Complexes.
        J. Mol. Biol. 2016; 428: 720-725
        • Adasme M.F.
        • Linnemann K.L.
        • Bolz S.N.
        • Kaiser F.
        • Salentin S.
        • Haupt V.J.
        • Schroeder M.
        PLIP 2021: expanding the scope of the protein–ligand interaction profiler to DNA and RNA.
        Nucleic Acids Res. 2021; 49: W530-W534
        • Pettersen E.F.
        • Goddard T.D.
        • Huang C.C.
        • Meng E.C.
        • Couch G.S.
        • Croll T.I.
        • Morris J.H.
        • Ferrin T.E.
        UCSF ChimeraX: Structure visualization for researchers, educators, and developers.
        Protein Sci. 2021; 30: 70-82
        • Ficarro S.B.
        • Browne C.M.
        • Card J.D.
        • Alexander W.M.
        • Zhang T.
        • Park E.
        • Mcnally R.
        • Dhe-Paganon S.
        • Seo H.-S.
        • Lamberto I.
        • Eck M.J.
        • Buhrlage S.J.
        • Gray N.S.
        • Marto J.A.
        Leveraging Gas-Phase Fragmentation Pathways for Improved Identification and Selective Detection of Targets Modified by Covalent Probes.
        Anal. Chem. 2016; 88: 12248-12254
        • Renuse S.
        • Madugundu A.K.
        • Jung J.H.
        • Byeon S.K.
        • Goldschmidt H.L.
        • Tahir R.
        • Meyers D.
        • Kim D.I.
        • Cutler J.
        • Kim K.P.
        • Wu X.
        • Huganir R.L.
        • Pandey A.
        Signature Fragment Ions of Biotinylated Peptides.
        J. Am. Soc. Mass Spectrom. 2020; 31: 394-404
        • Manning G.
        • Plowman G.D.
        • Hunter T.
        • Sudarsanam S.
        Evolution of protein kinase signaling from yeast to man.
        Trends Biochem. Sci. 2002; 27: 514-520
        • Moorhead G.B.G.
        • De Wever V.
        • Templeton G.
        • Kerk D.
        Evolution of protein phosphatases in plants and animals.
        Biochem. J. 2009; 417: 401-409
        • Liu X.
        • Rossio V.
        • Thakurta S.G.
        • Flora A.
        • Foster L.
        • Bomgarden R.D.
        • Gygi S.P.
        • Paulo J.A.
        Fe3+-NTA magnetic beads as an alternative to spin column-based phosphopeptide enrichment.
        J. Proteomics. 2022; 260104561
        • Sweet S.M.M.
        • Creese A.J.
        • Cooper H.J.
        Strategy for the Identification of Sites of Phosphorylation in Proteins: Neutral Loss Triggered Electron Capture Dissociation.
        Anal. Chem. 2006; 78: 7563-7569
        • Jozsef L.
        • Tashiro K.
        • Kuo A.
        • Park E.
        • Skoura A.
        • Albinsson S.
        • Rivera-Molina F.
        • Harrison K.D.
        • Iwakiri Y.
        • Toomre D.
        • Sessa W.C.
        Reticulon 4 Is Necessary for Endoplasmic Reticulum Tubulation, STIM1-Orai1 Coupling, and Store-operated Calcium Entry.
        J. Biol. Chem. 2014; 289: 9380
        • Bateman L.A.
        • Nguyen T.B.
        • Roberts A.M.
        • Miyamoto D.K.
        • Ku W.M.
        • Huffman T.R.
        • Petri Y.
        • Heslin M.J.
        • Contreras C.M.
        • Skibola C.F.
        • Olzmann J.A.
        • Nomura D.K.
        Chemoproteomics-enabled covalent ligand screen reveals a cysteine hotspot in reticulon 4 that impairs ER morphology and cancer pathogenicity.
        Chem. Commun. 2017; 53: 7234-7237
        • Pathak G.P.
        • Shah R.
        • Kennedy B.E.
        • Murphy J.P.
        • Clements D.
        • Konda P.
        • Giacomantonio M.
        • Xu Z.
        • Schlaepfer I.R.
        • Gujar S.
        (2018) RTN4 Knockdown Dysregulates the AKT Pathway, Destabilizes the Cytoskeleton, and Enhances Paclitaxel-Induced Cytotoxicity in Cancers.
        Mol. Ther. 2019–2033; 26
        • Poole L.B.
        The basics of thiols and cysteines in redox biology and chemistry.
        Free Radic. Biol. Med. 2015; 80: 148-157
        • Elko E.A.
        • Cunniff B.
        • Seward D.J.
        • Chia S.B.
        • Aboushousha R.
        • Van De Wetering C.
        • Van Der Velden J.
        • Manuel A.
        • Shukla A.
        • Heintz N.H.
        • Anathy V.
        • Van Der Vliet A.
        • Janssen-Heininger Y.M.W.
        Peroxiredoxins and Beyond; Redox Systems Regulating Lung Physiology and Disease.
        Antioxid. Redox Signal. 2019; 31: 1070-1091
        • Ralat L.A.
        • Misquitta S.A.
        • Manevich Y.
        • Fisher A.B.
        • Colman R.F.
        Characterization of the complex of glutathione S-transferase pi and 1-cysteine peroxiredoxin.
        Arch. Biochem. Biophys. 2008; 474: 109-118
        • Fisher A.B.
        Peroxiredoxin 6: A Bifunctional Enzyme with Glutathione Peroxidase and Phospholipase A2 Activities.
        Antioxid. Redox Signal. 2011; 15: 831
        • Zhou S.
        • Lien Y.C.
        • Shuvaeva T.
        • Debolt K.
        • Feinstein S.I.
        • Fisher A.B.
        Functional interaction of glutathione S-transferase pi and peroxiredoxin 6 in intact cells.
        Int. J. Biochem. Cell Biol. 2013; 45: 401-407
        • Hashemy S.I.
        • Holmgren A.
        Regulation of the Catalytic Activity and Structure of Human Thioredoxin 1 via Oxidation and S-Nitrosylation of Cysteine Residues.
        J. Biol. Chem. 2008; 283: 21890-21898
        • Han S.
        Molecular dynamics simulations of thioredoxin with S-glutathiolated cysteine-73.
        Biochem. Biophys. Res. Commun. 2007; 362: 532-537
        • Casagrande S.
        • Bonetto V.
        • Fratelli M.
        • Gianazza E.
        • Eberini I.
        • Massignan T.
        • Salmona M.
        • Chang G.
        • Holmgren A.
        • Ghezzi P.
        • Herzenberg L.A.
        Glutathionylation of human thioredoxin: A possible crosstalk between the glutathione and thioredoxin systems.
        PNAS. 2002; 99: 9745-9749
        • Mao Z.
        • Yang X.
        • Mizutani S.
        • Huang Y.
        • Zhang Z.
        • Shinmori H.
        • Gao K.
        • Yao J.
        Hydrogen Sulfide Mediates Tumor Cell Resistance to Thioredoxin Inhibitor.
        Front. Oncol. 2020; 10: 252
        • Chia S.B.
        • Elko E.A.
        • Aboushousha R.
        • Manuel A.M.
        • van de Wetering C.
        • Druso J.E.
        • van der Velden J.
        • Seward D.J.
        • Anathy V.
        • Irvin C.G.
        • Lam Y.W.
        • van der Vliet A.
        • Janssen-Heininger Y.M.W.
        Dysregulation of the glutaredoxin/S-glutathionylation redox axis in lung diseases.
        Am. J. Physiol. - Cell Physiol. 2020; 318: C304-C327
        • Hashemy S.I.
        • Johansson C.
        • Berndt C.
        • Lillig C.H.
        • Holmgren A.
        Oxidation and S-Nitrosylation of Cysteines in Human Cytosolic and Mitochondrial Glutaredoxins: EFFECTS ON STRUCTURE AND ACTIVITY.
        J. Biol. Chem. 2007; 282: 14428-14436
        • Mi H.
        • Muruganujan A.
        • Huang X.
        • Ebert D.
        • Mills C.
        • Guo X.
        • Thomas P.D.
        Protocol Update for large-scale genome and gene function analysis with the PANTHER classification system (v.14.0).
        Nat. Protoc. 2019; 14: 703-721
        • Mi H.
        • Ebert D.
        • Muruganujan A.
        • Mills C.
        • Albou L.P.
        • Mushayamaha T.
        • Thomas P.D.
        PANTHER version 16: a revised family classification, tree-based classification tool, enhancer regions and extensive API.
        Nucleic Acids Res. 2021; 49: D394-D403
        • Wennerberg K.
        • Rossman K.L.
        • Der C.J.
        The Ras superfamily at a glance.
        J. Cell Sci. 2005; 118: 843-846
        • Mukhopadhyay S.
        • Vander Heiden M.G.
        • McCormick F.
        The metabolic landscape of RAS-driven cancers from biology to therapy.
        Nat. Cancer. 2021; 2: 271-283
        • Simanshu D.K.
        • Nissley D.V.
        • McCormick F.
        RAS Proteins and Their Regulators in Human Disease.
        Cell. 2017; 170: 17-33
        • Aaron Hobbs G.
        • Zhou B.
        • Cox A.D.
        • Campbell S.L.
        Rho GTPases, oxidation, and cell redox control.
        Small GTPases. 2014; 5https://doi.org/10.4161/sgtp.28579
        • Cherfils J.
        • Zeghouf M.
        Regulation of Small GTPases by GEFs, GAPs, and GDIs.
        Physiol Rev. 2013; 93: 269-309
        • Hodge R.G.
        • Ridley A.J.
        Regulating Rho GTPases and their regulators.
        Nat. Rev. Mol. Cell Biol. 2016; 17: 496-510
        • Hurst M.
        • McGarry D.J.
        • Olson M.F.
        Rho GTPases: Non-canonical regulation by cysteine oxidation.
        BioEssays. 2022; 442100152
        • Pylypenko O.
        • Hammich H.
        • Yu I.M.
        • Houdusse A.
        Rab GTPases and their interacting protein partners: Structural insights into Rab functional diversity.
        Small GTPases. 2018; 9: 22-48
        • Tao G.Z.
        • Zhou Q.
        • Strnad P.
        • Salemi M.R.
        • Young M.L.
        • Omary M.B.
        Human Ran Cysteine 112 Oxidation by Pervanadate Regulates Its Binding to Keratins.
        J. Biol. Chem. 2005; 280: 12162-12167
        • Savitski M.M.
        • Reinhard F.B.M.
        • Franken H.
        • Werner T.
        • Savitski M.F.
        • Eberhard D.
        • Molina D.M.
        • Jafari R.
        • Dovega R.B.
        • Klaeger S.
        • Kuster B.
        • Nordlund P.
        • Bantscheff M.
        • Drewes G.
        Tracking cancer drugs in living cells by thermal profiling of the proteome.
        Science (80-. ). 2014; 346https://doi.org/10.1126/science.1255784
        • Ostermeier C.
        • Brunger A.T.
        Structural Basis of Rab Effector Specificity: Crystal Structure of the Small G Protein Rab3A Complexed with the Effector Domain of Rabphilin-3A.
        Cell. 1999; 96: 363-374
        • Han J.
        • Weisbrod R.M.
        • Shao D.
        • Watanabe Y.
        • Yin X.
        • Bachschmid M.M.
        • Seta F.
        • Janssen-Heininger Y.M.W.
        • Matsui R.
        • Zang M.
        • Hamburg N.M.
        • Cohen R.A.
        The redox mechanism for vascular barrier dysfunction associated with metabolic disorders: Glutathionylation of Rac1 in endothelial cells.
        Redox Biol. 2016; 9: 306-319
        • Navarro-Lérida I.
        • Sánchez-Perales S.
        • Calvo M.
        • Rentero C.
        • Zheng Y.
        • Enrich C.
        • Del Pozo M.A.
        A palmitoylation switch mechanism regulates Rac1 function and membrane organization.
        EMBO J. 2012; 31: 534-551
        • Thomas P.D.
        • Campbell M.J.
        • Kejariwal A.
        • Mi H.
        • Karlak B.
        • Daverman R.
        • Diemer K.
        • Muruganujan A.
        • Narechania A.
        PANTHER: a library of protein families and subfamilies indexed by function.
        Genome Res. 2003; 13: 2129-2141