TY - JOUR
AU - Skipp,, Paul
AB - Abstract The risk of contact sensitization is a major consideration in the development of new formulations for personal care products. However, developing a mechanistic approach for non-animal risk assessment requires further understanding of haptenation of skin proteins by sensitizing chemicals, which is the molecular initiating event causative of skin sensitization. The non-stoichiometric nature of protein haptenation results in relatively low levels of modification, often of low abundant proteins, presenting a major challenge for their assignment in complex biological matrices such as skin. Instrumental advances over the last few years have led to a considerable increase in sensitivity of mass spectrometry (MS) techniques. We have combined these advancements with a novel dual-labeling/LC-MSE approach to provide an in-depth direct comparison of human serum albumin (HSA), 2,4-dinitro-1-chlorobenzene (DNCB), 5-chloro-2-methyl-4-isothiazolin-3-one (MCI), trans-cinnamaldehyde, and 6-methyl coumarin. These data have revealed novel insights into the differences in protein haptenation between sensitizers with different reaction mechanisms and sensitizing potency; the extreme sensitizers DNCB and MCI were shown to modify a greater number of nucleophilic sites than the moderate sensitizer cinnamaldehyde; and the weak/non-sensitizer 6-methyl coumarin was restricted to only a single nucleophilic residue within HSA. The evaluation of this dual labeling/LC-MSE approach using HSA as a model protein has also demonstrated that this strategy could be applied to studying global haptenation in complex mixtures of skin-related proteins by different chemicals. contact allergy, GeLC-MS/MS, proteomics, sensitizer, skin Skin sensitization, which leads to the development of allergic contact dermatitis (ACD), is an immune response to commonly harmless chemicals causing significant morbidity without any benefit to those affected (McFadden et al., 2013). It was first proposed over 75 years ago that the key event in the induction of skin sensitization is the covalent modification (haptenation) of skin proteins by low molecular weight chemicals (haptens), which are too small to be recognized by an immune system (Landsteiner and Jacobs, 1935). This phenomenon has been studied extensively, utilizing a variety of techniques and models, including peptides and relevant proteins to study haptenation by sensitizers (e.g., Aleksic et al., 2007, 2008; Alvarez-Sanchez et al., 2004a,b ; Gerberick et al., 2007). Despite different study designs, these studies conclude that binding of sensitizing chemicals to protein nucleophiles is selective and governed by three main factors: electrophilicity of the sensitizer, nucleophilicity of the target, and steric constraints. Most sensitizers are electrophilic, reacting predominantly with nucleophilic side chains of lysine and cysteine, less commonly tyrosine, histidine, and arginine (Ahlfors et al., 2003). Many intra and extracellular events can be triggered by sensitizer reactivity. For example, reactivity drives glutathione conjugation as well as activation of the Nrf2 pathway and thus induction of phase II metabolizing enzymes via covalent binding to Cys residues on Keap1 protein (Migdal et al., 2013). An insight into the dynamic interplay between the detoxification events leading to the removal of an electrophilic chemical and haptenation of proteins, followed by elicitation of an immune response, is sorely missed. Thus an understanding of qualitative and particularly quantitative aspects of skin protein haptenation is one of the major gaps in advancing our knowledge of the adverse outcome pathway of induction (and elicitation) of skin sensitization (OECD, 2012). Determination of sites of protein modification in mixtures represents a considerable analytical challenge. Technical and instrumental limitations have to date restricted detailed investigations of protein haptenation in complex protein mixtures and tissues including human skin. In skin sensitization research, in parallel to developments elsewhere (reviewed by Liebler, 2008), methods for studying protein haptenation in mixtures range from immunochemistry-based studies, utilizing antibodies specific to the adduct(s) formed (e.g., Elahi et al., 2004), to proteomics and Nuclear Magnetic Resonance (NMR)-based techniques (e.g., Elbayed et al., 2013). Some investigators were able to utilize unique features of certain sensitizers (such as fluorescent adducts of monobromobimane) to pinpoint the amino acid site of haptenation in human skin (Simonsson et al., 2011); however, such techniques are only applicable to chemicals possessing those unique features. Further afield from sensitization, researchers have demonstrated ways to overcome the most critical issue of low abundance of modified proteins in cell/tissue extracts. A variety of techniques have been used, ranging from 2D gel-based separations using radiolabeled probes (e.g., Tzouros and Pahler, 2009), affinity capture, and shotgun proteomics strategies utilizing biotin-linked electrophiles (e.g., Dennehy et al., 2006), to post-labeling click chemistry approaches (Vila et al., 2008). Although very useful, none of the above approaches are ideal and raise additional issues, for example cost of synthesis, radiolabel stability, and the influence of bulky linker and affinity tags on target selectivity of an otherwise small electrophile structure. There is to date no globally applicable methodology to study protein haptenation in complex mixtures. Instrumental advances have led to a considerable increase in sensitivity of mass spectrometry (MS) techniques recently. Here we combine the gel-based approach GeLC-MS/MS (Schirle et al., 2003) with a novel dual-labeling/liquid chromatography-MSE approach (LC-MSE) to provide an in-depth direct comparison of human serum albumin (HSA) haptenation by classical sensitizers. HSA is well characterized, prevalent in skin, and has been previously used in binding studies (Curry et al., 1998). A number of studies have also shown specific T cell responses after stimulation with haptenated HSA; including via a stable HSA-penicillin G complex (Brander et al., 1995); through the presence of dinitrophenyl (DNP) adducts after modification with the extreme sensitizer 2,4-dinitrobenzesulfonic acid (DNBS) (Dietz et al., 2010); via p-phenylenediamine (PPD) modification at position Cys 34 (Jenkinson et al., 2010); or by transporting nickel(II) directly to TCR/MHC conjugates (Thierse et al., 2004), suggesting that HSA may play a role as a carrier protein mediating T cell activation in contact allergy. We have studied HSA haptenation by well-understood sensitizers dinitrochlorobenzene (DNCB), cinnamaldehyde, and 5-chloro-2-methylisothiazol-3-one (MCI). In addition, we investigated haptenation by 6-methyl coumarin, which has been classed as a non-sensitizer in the murine local lymph node assay (Ashby et al., 1995). Our initial studies have shown that haptenation of proteins is non-stoichiometric in nature, presenting challenges for their analysis. By taking advantage of the MSE data independent mode of data acquisition (Silva et al., 2006), and combining this with the dual labeling approach, we were able to pinpoint low abundant modified peptides using an isotopic precursor ion signature and confirm the peptide identity and amino acid site of haptenation using product ion fragmentation data. MATERIALS AND METHODS Test chemicals DNCB (99% purity; MW 202.55 Da) was obtained from Sigma, UK, and DNCB-D3 (99% purity; MW 205.57Da) was obtained from QMX Laboratories. Trans-cinnamaldehyde (CA) (99% purity; MW 132.16 Da) and 6-methyl coumarin (99% purity; MW 160.17 Da) were obtained from Sigma-Aldrich and trans-cinnamaldehyde-D5 (98% purity; MW 137.12 Da) and 6-methyl coumarin-D3 (99% purity; MW 163.15 Da) were custom synthesized by Quotient Amersham Radiochemicals. 5-chloro-2-methyl-4-isothiazolin-3-one (MCI) (MW 149.60 Da) and C13-labeled MCI (MW 150.8 Da) were synthesized and kindly donated by Professor Jean-Pierre Lepoittevin and Dr Elena Gimenez Arnau, Labarotoire de Dermatochimie, Strasbourg. Exact locations of the stable isotopes for each chemical are shown in Table 1. Structures, Position of Stable Isotope, Potency Category (Including EC3 Value, as Derived from the Local Lymph Node Assay), Δ Mass (Da) Expected Following Haptenation, Reactivity Domain, Variable Modifications of Sensitizers for Database Searching of MS Data Within PLGS TABLE 1. Structures, Position of Stable Isotope, Potency Category (Including EC3 Value, as Derived from the Local Lymph Node Assay), Δ Mass (Da) Expected Following Haptenation, Reactivity Domain, Variable Modifications of Sensitizers for Database Searching of MS Data Within PLGS Open in new tab TABLE 1. Structures, Position of Stable Isotope, Potency Category (Including EC3 Value, as Derived from the Local Lymph Node Assay), Δ Mass (Da) Expected Following Haptenation, Reactivity Domain, Variable Modifications of Sensitizers for Database Searching of MS Data Within PLGS Open in new tab HSA sensitizer modification To study the effect of DNCB haptenation with increasing concentration and time, HSA (fraction V, >97% purity, A9511 (lot no. 107K7560), Sigma) was solubilized in PBS (pH 7.4) to a final concentration of 1 mg/ml, DNCB and deuterated DNCB were solubilized in 100% Dimethyl sulphoxide (DMSO). HSA was incubated with either DNCB or deuterated DNCB at a range of concentrations (1500, 750, 150, 15, 7.5μM; final concentration of DMSO < 1%) for either 1, 6, 24, or 192 h (8 days) at 37°C. The relative amount of protein modification was measured by absorbance at 360 nm for each dose of DNCB at each time point (Kesner et al., 1967). Measurements were background corrected with background calculated as mean of the absorbance of DNCB + PBS control at 360 nm minus mean of the absorbance of HSA + PBS control at 360 nm (n = 3). To maximize the amount of haptenated peptides and determine the differences in haptenation with a range of chemicals, stock solutions of sensitizers were prepared in 100% DMSO (for DNCB) or 100% ethanol (for cinnamaldehyde, MCI, and 6-methyl-coumarin) containing 50%, by molar concentration, of unlabeled sensitizer and 50% stable isotope labeled sensitizer. HSA (fraction V, >95% purity, 126658 (lot no. D00095847), Calbiochem, UK) was solubilized in 0.1-M TEAB (pH 8.0) at 1 mg/ml prior to modification with a 1:100 molar excess of sensitizer to protein (control samples were prepared in 0.1-M Triethylammonium bicarbonate (TEAB) (pH 8.0) with the addition of 0.2% of the relevant solvent) and incubated at 37°C. Aliquots of sensitizer-modified (or control) HSA were taken at 24 h, 2 weeks, and 4 weeks. Free chemical and salts were removed by resolving the samples through an SDS polyacrylamide gel for 45 min. Assessment of the reducing capacity of DNCB Bodipy-FL-cysteine (5μM in PBS, pH 7.4; or 50-mM sodium acetate + 100-mM NaCl, pH 4.5; Life Science Technologies, Paisley, UK) was incubated for 1 h at 37°C with varying concentrations of DNCB (2.5, 5, 50, 250, 500, and 1000μM) or Dithiothreitol (DTT) (25, 250μM, 2.5, 5, 10, 15mM) as a positive control. The fluorescence of the treated Bodipy-FL-cysteine was measured (excitation λ 485 nm, emission λ 530 nm) to determine the extent of reduction. GeLC-MS/MS Modified HSA and control samples (20–40 ug) were denatured in final sample buffer (NuPAGE LDS Sample Buffer, Life Technologies, Paisley, UK) containing 10-mM DTT at 70°C for 10 min. The denatured samples were resolved using a NuPage 4–12% Bis-Tris gel (Life Technologies) at 200 V for approximately 45 min. Following staining with Coomassie Blue, the large band at approximately 66 kDa was excised and subjected to in situ trypsin digestion using the method of Shevchenko et al. (1996). The resulting peptides were extracted and ∼1–2 μg of sample loaded onto a reverse phase trap column (Xbridge BEH C18 NanoEase column, 5 μm, 300 μm x 50 mm, Waters, Manchester, UK) at a trapping rate of 5 μl/min and washed for 10 min with buffer A prior to the analytical nano-LC separation using a C18 Reverse phase column (BEH C18, 1.7 μm, 150 mm x 75 μm, Waters). The eluted peptides were fractionated over a 37-min continuous gradient from 1% acetonitrile + 0.1% formic acid up to 80% acetonitrile + 0.1% formic acid, at a flow rate of 300 nl/min. Eluted samples were sprayed directly into a Global Ultima Q-ToF mass spectrometer (Waters), fitted with a nanolockspray source and operating in MSE mode. Data were acquired from 50 to 1990 m/z using alternate low and high collision energy (CE) scans, where the low CE was set to 5 V and elevated CE was set to 30 V. The lock mass Glu-fibrinopeptide, ((M+2H)+2, m/z = 785.8426) was infused at a concentration of 500 fmol/μl with a flow rate of 250 nl/min and acquired every 13 s. Database searches The raw mass spectra were processed using ProteinLynx Global Server Ver 2.4 (Waters) and the data processed to generate reduced charge state and deisotoped precursor and associated product ion peak lists. These peak lists were searched against the HSA sequence (obtained from UniProt 03/2010). A maximum of two missed cleavages was allowed for tryptic digestion and the variable modification was set to contain oxidation of methionine, carboxyamidomethylation of cysteine, and sensitizer-specific haptenation(s) as detailed in Table 1. Precursor ion and product ion mass tolerances were set at 150 ppm and 0.2 Da, respectively, and the false positive rate was set at 2%. Bioinformatic analysis As the modification of a protein by a sensitizer is non-stoichiometric, modified peptide ions are lower in abundance and often found within the “noise” of a mass spectrum. This combined with the increased search space resulting from the addition of modifications to the database searches leads to a large number of false positive assignments. In this study, we have employed a dual isotope labeling strategy to enable us to eliminate false positives as follows. Following database searching, precursor ion peak pairs were extracted from the Peptide 3D files generated from processing raw data using PLGS based on the following criteria; ion pairs with a fixed mass difference (corresponding to the number of stable isotopes incorporated into the labeled sensitizer), with similar ion intensity and within a retention time window of 1 min. Extracted peptide pairs were correlated with modified peptide masses identified after database searching using m/z and retention time. Spectra were subsequently manually inspected using MassLynx 4.1 and the amino acid site of modification determined, where possible. The structure of HSA was imported from Protein Data Bank (PDB, http://www.pdb.org/pdb/home/home.do; structure 1AO6) into the molecular viewer and 3D molecular editor software PyMOL version 0.99 for editing, enabling the modified amino acid residues to be highlighted within the structures. RESULTS Initial haptenation studies comparing the intensities of haptenated peptides within complex samples revealed their low abundance compared with unmodified peptides (data not shown). To reduce the bias toward highly abundant peptide species and improve the ability to detect and identify low abundant haptenated peptides, we replaced the Data Dependent Acquisition commonly used for these types of studies, with the data-independent mode of acquisition, MSE (Geromanos et al., 2009; Michalski et al., 2011). Time points for the investigation were chosen by monitoring the rate of binding of DNCB to HSA using ultraviolet (UV) absorbance at 360 nm at four different time points (Fig. 1). The observed levels of haptenation increased as a function of time and concentration, adjusting for a low level of HSA autofluorescence. FIG. 1. Open in new tabDownload slide The rate of HSA haptenation by DNCB monitored by the increase in absorbance at 360 nm as a function of incubation time. Data from triplicate absorbance readings (n = 3) mean A360 of sample minus background, with background calculated as mean A360 of DNCB + PBS control minus mean A360 of HSA + PBS control. FIG. 1. Open in new tabDownload slide The rate of HSA haptenation by DNCB monitored by the increase in absorbance at 360 nm as a function of incubation time. Data from triplicate absorbance readings (n = 3) mean A360 of sample minus background, with background calculated as mean A360 of DNCB + PBS control minus mean A360 of HSA + PBS control. We used the dual labeling approach as a way of eliminating false positives, whereby samples were haptenated with a 50:50 molar ratio of unlabeled and stable isotope-labeled sensitizer; haptenated peptides were identified by two peptide clusters, the first cluster relating to the modified peptide and the second cluster at a fixed mass difference relating to the peptide modified with the stable isotope-labeled sensitizer. Figure 2 illustrates incorporation of MSE and stable isotope labeling into the sample preparation and MS analysis workflow. FIG. 2. Open in new tabDownload slide Protein samples are haptenated using a 50:50 molar ratio of unlabeled and stable isotope-labeled sensitizers (a) prior to removal of excess sensitizer by resolving the protein through a 4–12% Bis/Tris polyacrylamide gel (b). The protein band/s of interest are excised and subjected to in gel reduction, alkylation, and trypsin digestion (c). Extracted peptides are injected onto a C18 nano LC column and eluted over a 45-min gradient (0–40% acetonitrile) on-line to Global Ultima Q-ToF mass spectrometer (Waters), operating in MSE mode. The presence of peptide pairs at the same retention time separated by a delta mass corresponding to the number of incorporated stable isotopes of the chemical tested confirms the presence of a modified peptide. Analysis of the fragmentation spectra allows the identification of the modified amino acid (d). FIG. 2. Open in new tabDownload slide Protein samples are haptenated using a 50:50 molar ratio of unlabeled and stable isotope-labeled sensitizers (a) prior to removal of excess sensitizer by resolving the protein through a 4–12% Bis/Tris polyacrylamide gel (b). The protein band/s of interest are excised and subjected to in gel reduction, alkylation, and trypsin digestion (c). Extracted peptides are injected onto a C18 nano LC column and eluted over a 45-min gradient (0–40% acetonitrile) on-line to Global Ultima Q-ToF mass spectrometer (Waters), operating in MSE mode. The presence of peptide pairs at the same retention time separated by a delta mass corresponding to the number of incorporated stable isotopes of the chemical tested confirms the presence of a modified peptide. Analysis of the fragmentation spectra allows the identification of the modified amino acid (d). Modified peptides were confirmed where two peptide clusters of fixed Δmass were observed in the MS spectrum. The MS/MS spectrum of the modified precursor ion was subsequently analyzed to determine the exact site of haptenation. An example of MS and MS/MS spectra generated is shown in Supplementary Data. A total of 34 residues were confirmed to be haptenated by DNCB, 33 by MCI, and nine by cinnamaldehyde (Supplementary Data). No confirmed modifications were observed for 6-methyl coumarin. For all chemicals tested, we detected a number of potential sites of haptenation which unfortunately could not be confirmed as the fragmentation spectra were of insufficient quality to determine the haptenation location. For example, the most prominent signal for 6-methyl coumarin haptenation was for the peptide ADDKETCFAEEGKK, which included two peptide clusters of 1.5 Da apart. Three lysines, Lys 564, Lys 573, and Lys 574, are contained within the peptide and are likely to represent one of the potentially haptenated residues (Supplementary Data). Some common features were observed when comparing data from different chemicals. DNCB and MCI both preferentially haptenate Lys (Supplementary Data), possibly due to higher proportion of Lys in HSA compared with other nucleophilic residues. Interestingly, cinnamaldehyde showed high affinity for haptenating Arg residues. Double cinnamaldehyde adducts have previously been observed on the guanidine moiety of Arg side chains (Aleksic et al., 2009), but this was not investigated in the present study due to complexity. The majority of DNCB-haptenated residues were observed throughout the time course and as early as 24 h suggesting that the same residues susceptible to haptenation continue to be haptenated over time. However, certain residues only appear haptenated by DNCB at later time points (e.g., His 3, Lys 262, and Tyr 263). Similar observations were made for MCI (where His 3, Lys 4, Lys 106, and Lys 181 were found haptenated after 24 h) and cinnamaldehyde (where His 128 was haptenated after 24 h). This may possibly be due to a change in structure of the protein molecule by initial haptenation, thus making new nucleophilic sites susceptible to further haptenation. Several residues appeared modified at 24 h by DNCB, but could not be detected at the later time points (Lys 12, Cys 34, Lys 64, Cys 75, Lys 93, Cys 124, Cys 177, Lys 432, and Lys 545) (Fig. 3). On further inspection, evidence of possible further modifications such as oxidation and methylation of these modified HSA peptides was observed which would obscure these signals from database searches. FIG. 3. Open in new tabDownload slide Schematic drawing of the secondary structure of a single molecule of HSA (downloaded from PDB, 1AO6), front and back. Modified nucleophilic residues are highlighted and shown as stick structures, the remaining unmodified nucleophilic residues are highlighted only. This figure was generated using PyMol. FIG. 3. Open in new tabDownload slide Schematic drawing of the secondary structure of a single molecule of HSA (downloaded from PDB, 1AO6), front and back. Modified nucleophilic residues are highlighted and shown as stick structures, the remaining unmodified nucleophilic residues are highlighted only. This figure was generated using PyMol. Four HSA lysines are commonly targeted by all tested haptens. These include Lys190, Lys212, Lys475, and Lys574 (Fig. 4). Conversely, certain nucleophilic residues are exclusively haptenated by a single sensitizer. For example, DNCB exclusively haptenates Cys34, Lys64, Cys75, Lys93, Cys124, Cys177, Tyr263, Cys392, Tyr411, and Lys413; MCI exclusively haptenates Lys51, Lys159, Tyr161, Lys276, Tyr319, Lys323, Lys538, and Lys564; and cinnamaldehyde exclusively haptenates His128, Arg186, Arg218, and Arg428 (Fig. 4). FIG. 4. Open in new tabDownload slide Circular representation of the sequence of HSA showing which nucleophiles were haptenated after exposure with the test chemicals DNCB, MCI, and cinnamaldehyde. The four amino acids that were found to be haptenated by all three chemicals are outlined with a rectangular box. FIG. 4. Open in new tabDownload slide Circular representation of the sequence of HSA showing which nucleophiles were haptenated after exposure with the test chemicals DNCB, MCI, and cinnamaldehyde. The four amino acids that were found to be haptenated by all three chemicals are outlined with a rectangular box. Contrastingly, there are three regions of HSA where no haptenation sites were found with any of the test chemicals used (residues 287–316, 437–472, and 477–509). Although these regions contain nucleophilic residues, it is likely that the tertiary structure of HSA prevented haptenation as the nucleophilic side chains are facing inward toward the center of the molecule (Fig. 5). FIG. 5. Open in new tabDownload slide Schematic drawing of the secondary structure of a dimer of HSA (downloaded from PDB, 1AO6) showing the regions of the HSA molecule that are not modified by any of the chemicals tested. The unmodified regions 287–316, 437–472, and 477–509 are highlighted in orange on the secondary structure of both HSA molecules. Nucleophilic residues within these unmodified regions are shown in pink on the right-hand side molecule. The three regions are circled with the amino acid sequence of each region stated next to the relevant areas. FIG. 5. Open in new tabDownload slide Schematic drawing of the secondary structure of a dimer of HSA (downloaded from PDB, 1AO6) showing the regions of the HSA molecule that are not modified by any of the chemicals tested. The unmodified regions 287–316, 437–472, and 477–509 are highlighted in orange on the secondary structure of both HSA molecules. Nucleophilic residues within these unmodified regions are shown in pink on the right-hand side molecule. The three regions are circled with the amino acid sequence of each region stated next to the relevant areas. DNCB is probably the most studied skin sensitizer and has previously been shown to bind to the Lys, His, Tyr, and Cys via nucleophilic substitution (SNAr), resulting in the addition of a dinitrophenyl adduct with a mass of +166 Da (Aleksic et al., 2007). In our study, DNCB modifies a large number of available nucleophilic residues on HSA (Fig. 3). The degree of dinitrophenylation of HSA, as determined by measuring the absorbance at 360 nm with a range of DNCB concentrations over time, suggests that DNCB concentration has a significant effect on the overall amount of modified protein (Fig. 1). Because no plateau in the rate of modification (Fig. 1) was observed, it is likely that saturation of the protein has not yet occurred by the end of the 1-week time course studied. This may suggest that DNCB haptenation causes exposure of more nucleophilic sites due to structural changes within HSA. In support of this, a number of studies have shown changes in the secondary and tertiary structure of the protein in response to temperature (Wetzel et al., 1980), pH (Bhattacharya et al., 2011), chemical denaturation (Santra et al., 2005), ligand binding (Bertucci and Domenici, 2002), and covalent modification by chemicals (Anraku et al., 2001). In total, 34 amino acids were confirmed to be haptenated by DNCB. This is in contrast to previous work where it was considered that DNCB modification was limited to only 10 sites (Aleksic et al., 2007). The differences in these findings are most likely due to the increase in sensitivity of the instrumentation, the mode of data acquisition employed (MSE), and the use of stable isotopes. Interestingly, three residues reported as DNCB-haptenated by Aleksic et al. (2007), N-term, His 9, Tyr 140, were absent from these data, although modifications were found within the same peptides at positions His 3, Lys 4, and Lys 137. This could potentially be explained by differences in incubation conditions, particularly because the Aleksic study used a high proportion of organic solvent in the incubation buffer, which could have effects on the conformation of the protein and potentially the reactivity toward DNCB (Griebenow and Klibanov, 1996). The modification of Cys residues that are reported to be S-S bridged in HSA was repeatedly observed only with DNCB and was confirmed at multiple time points except time zero. It is therefore unlikely that these were generated as a consequence of sample processing. We assessed the reducing capacity of DNCB by incubation with bodipy-FL-cysteine, which fluoresces when reduced, and compared this with the strong reducing agent DTT. No fluorescence was detected after treatment with DNCB confirming that DNCB is unlikely to be capable of reducing the disulphide bridges in HSA. These cysteine modifications were observed with HSA from two sources (Sigma and Calbiochem) suggesting that this is not preparation specific and may be as a result of low levels of free thiols that may have become available for modification, for example through the liberation of a free thiol via release of the nitroso group from an S-nitrosylated cysteine residue. Alternatively, DNCB may be unstable and mobile allowing modification of thiols liberated during the reduction step of the in-gel digestion procedure. Further work will be required to determine why we observe modifications of these cysteines by DNCB which are thought to exist only in disulphide bonds. We confirmed a total of 33 residues haptenated by MCI. The majority of haptenated residues were Lys and others included His and Tyr. Cys 34, the only Cys residue in HSA not involved in an S-S bridge was not found haptenated by MCI. A study by Alvarez-Sanchez et al. (2004a) found that MCI exclusively modified His and Lys residues and suggested that a lack of cysteine modification may have been due to the limited accessibility of MCI for Cys34. It is likely that MCI requires an initial interaction with a thiol for the reaction to progress through to generation of amide (+99) and thioamide (+115) adducts (Alvarez-Sanchez et al., 2004b), which were the two most commonly observed adducts of MCI in this study (+99 and +115 Da, respectively). This is possibly accomplished by reactivity with glutathione in vivo, however, it appears that a single Cys residue available in HSA is sufficient for this purpose. The numbers of sites haptenated by MCI are comparable to that by DNCB, both of which are extreme sensitizers. Cinnamaldehyde is a moderate sensitizer and reacts with protein nucleophiles either by formation of a Schiff base with Lys residues, resulting in a mass increase of +114 Da, or by Michael addition to Cys residues, resulting in a mass increase of +132 Da (Majeti and Suskind, 1977). Both types of adduct were observed with haptenation of fewer Lys, His, and Arg residues when compared with DNCB and MCI. The data do not show a preference for either mechanism of haptenation. This is in contrast to the previous assignment of cinnamaldehyde as a Michael acceptor as a preferred reaction mechanism (Roberts et al., 2007). 6-methylcoumarin, a synthetic fragrance material, has been defined both as a non-sensitizer in the murine LLNA and as a photo-contact sensitizer (Ashby et al., 1995; Kato et al., 1985; Maguire and Kaidbey, 1982). Because this compound was not photoactivated for the purposes of this study, 6-methylcoumarin is defined as a weak/non-sensitizer. However, low levels of reactivity have been shown in peptide depletion assays with Cys and Lys without photo-activation (Aleksic et al., 2009). In our experiments, no reactivity was observed with HSA after 24 h and 2-week exposure, supporting this compound as a weak/non-sensitizer. Haptenation of Lys 574 observed after 4-week exposure remains putative due to insufficient quality of fragmentation spectra to assign the amino acid site of modification (Supplementary Data). DISCUSSION A cascade of events triggered by covalent modification of skin proteins (haptenation) ultimately lead to the generation of the adaptive immune response termed skin sensitization. The resulting skin condition, ACD, has an appreciable effect on the quality of life and it affects a significant number of individuals. Nonetheless little is known about the true nature of this key molecular initiating event. Due to the experimental and analytical complexity, our current knowledge of haptenation mainly comes from studies utilizing surrogates such as model peptides or isolated single proteins (e.g., butylamine, propanethiol, N-acetyl-cysteine, etc. (Alvarez-Sanchez et al., 2003; Franot et al., 1994; Meschkat et al., 2001a), glutathione (Aptula et al., 2006), synthetic peptides (Aleksic et al., 2009; Alvarez-Sanchez et al., 2004b; Gerberick et al., 2004; Natsch and Gfeller 2008; Natsch et al., 2007) to single isolated model proteins (Aleksic et al., 2007, 2008; Alvarez-Sanchez et al., 2004a; Meschkat et al., 2001b)). Thus, much is known about the similarities of individual haptens or groups of haptens with related reactivity mechanisms. So far, the proteome that is being haptenated, i.e., the human skin, was largely overlooked. Advances in proteomics and instrument sensitivity as well as greatly increased sample throughput and software capabilities have now allowed detailed investigations of the skin proteome (Parkinson et al., 2014), but the understanding of the extent and specificity of haptenation in a complex proteome is still lacking. To further our knowledge of haptenation, it is necessary to investigate mechanisms and specificity of covalent protein modifications in cells and skin tissue by overcoming the technical and analytical challenge that this possesses. In the present study, improvements in the sensitivity, resolution, and modes of acquisition of modern MS combined with a novel isotopic labeling strategy have allowed us to identify low abundant haptenated peptides on a single model protein despite the non-stoichiometric nature of haptenation. In comparison to previous similar studies, major differences were noted and attributed to improved technique and instrumental sensitivity. This novel approach should enable detailed investigation of haptenation events in complex protein mixtures, cell lines, and ultimately ex vivo topically treated human skin. Understanding the mechanisms and extent of haptenation in human skin will be essential to understanding the immunogenicity of haptenated proteins. Misleading conclusions can potentially be drawn from using very simple models, such as nucleophiles and nucleophile containing peptides. For example, cinnamaldehyde is an α, β-unsaturated aldehyde that can react either via Schiff base formation or Michael addition or both. Theoretically, Michael addition reaction is the preferred reaction mechanism for cinnamaldehyde (Roberts et al., 2007), however, both types of adducts were observed when using synthetic peptides (Aleksic et al., 2009). In isolation, cinnamaldehyde reacts with Cys, Lys, His, and Arg residues on single nucleophile peptides. However, our data show that cinnamaldehyde appears to preferentially haptenate Lys and Arg residues on HSA, which was somewhat unexpected. Basic amino acids Lys and Arg are higher in abundance in the human skin proteome when compared with Cys (Parkinson et al., 2014); therefore, it is more likely that immunogenic entities generated by cinnamaldehyde haptenation will occur on Lys- and Arg-rich proteins, whereas Cys haptenation (e.g., to Cys of glutathione) may result in detoxification. It is, however, difficult to speculate about the exact immunological or toxicological implications of the modifications observed on a model protein. Beyond description of Cys54 of keratin 5 as a major (but not the only) target for monobromobimane (Simonsson et al., 2011), little evidence exists to suggest which proteins or indeed which nucleophilic residues may be targeted by haptens and how specific haptenation events could influence immunogenicity. It is likely that chemicals capable of modifying a variety of nucleophilic side chains may be more successful in generating a larger number and variety of antigens than those reactive to a single nucleophile type. Conversely, a single modification of a key residue may still lead to a T cell response, in preference to more abundant modifications, for example proliferation of PPD-specific T cells was observed after stimulation of HSA modified at Cys 34 (Jenkinson et al., 2010). This remains a gap in our knowledge, thus understanding the variety and abundance of haptenation in human skin is the first step in elucidating molecular initiating events in skin allergy. Our future aim is to use the stable isotope methodology developed here to study haptenation in relevant complex protein mixtures including cell lines and ex vivo tissue. The use of data-independent MS acquisition, more sensitive MS instrumentation, and the dual isotope labeling strategy implemented in this study has revealed more about the modification of HSA by chemical sensitizers than was previously known. The strength of these approaches means that their application in larger studies looking at the global modification of complex mixtures of skin-related proteins by different chemicals is now possible. Similarly, this approach is likely to find utility in drug and drug metabolite binding studies. In particular quantitative studies of skin protein haptenation may enable estimation of the levels of modified protein which is thought to be a key metric in the induction of skin sensitization (MacKay et al., 2013; Maxwell et al., 2014). Increasing our understanding of the skin proteome, potential target protein(s) and the immunogenicity of the covalent modifications will ultimately enable us to better interpret reactivity data obtained from reactivity studies using simple model nucleophiles. FUNDING SEAC, Unilever plc (CH-2208-0181); BBSRC-Unilever Industrial Case Studentship (BB/G529832/1); Science Research Infrastructure Fund; Wessex Medical Trust. MCI and C13-labeled MCI were obtained from Professor Jean-Pierre Lepoittevin and Dr Elena Gimenez-Arnau, Laboratoire de Dermatochimie, Universite Louis Pasteur, Strasbourg, France. Conflict of interest: This study was funded by SEAC, Unilever plc. contract number CH-2008-0181. M.A. and R.C. were employees of Unilever and the salary of E.P. was funded by Unilever at the time of the study. P.B. was funded by a BBSRC-Unilever Industrial Case Studentship, BB/G529832/1. The funders did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. There are no patents, products in development or marketed products to declare. REFERENCES Ahlfors S. R. Sterner O. Hansson C. Reactivity of contact allergenic haptens to amino acid residues in a model carrier peptide, and characterization of formed peptide-hapten adducts Skin Pharmacol. Appl. Skin Physiol. 2003 16 59 68 Google Scholar Crossref Search ADS PubMed WorldCat Aleksic M. Pease C. K. Basketter D. A. Panico M. Morris H. R. Dell A. Investigating protein haptenation mechanisms of skin sensitisers using human serum albumin as a model protein Toxicol. In Vitro 2007 21 723 733 Google Scholar Crossref Search ADS PubMed WorldCat Aleksic M. Pease C. K. Basketter D. A. Panico M. Morris H. R. Dell A. Mass spectrometric identification of covalent adducts of the skin allergen 2,4-dinitro-1-chlorobenzene and model skin proteins Toxicol. In Vitro 2008 22 1169 1176 Google Scholar Crossref Search ADS PubMed WorldCat Aleksic M. Thain E. Roger D. Saib O. Davies M. Li J. Aptula A. Zazzeroni R. Reactivity profiling: Covalent modification of single nucleophile peptides for skin sensitization risk assessment Toxicol. Sci. 2009 108 401 411 Google Scholar Crossref Search ADS PubMed WorldCat Alvarez-Sanchez R. Basketter D. Pease C. Lepoittevin J. P. Studies of chemical selectivity of hapten, reactivity, and skin sensitization potency. 3. Synthesis and studies on the reactivity toward model nucleophiles of the 13C-labeled skin sensitizers, 5-chloro-2-methylisothiazol-3-one (MCI) and 2-methylisothiazol-3-one (MI) Chem. Res. Toxicol. 2003 16 627 636 Google Scholar Crossref Search ADS PubMed WorldCat Alvarez-Sanchez R. Basketter D. Pease C. Lepoittevin J. P. Covalent binding of the 13C-labeled skin sensitizers 5-chloro-2-methylisothiazol-3-one (MCI) and 2-methylisothiazol-3-one (MI) to a model peptide and glutathione Bioorg. Med. Chem. Lett. 2004a 14 365 368 Google Scholar Crossref Search ADS WorldCat Alvarez-Sanchez R. Divkovic M. Basketter D. Pease C. Panico M. Dell A. Morris H. Lepoittevin J. P. Effect of glutathione on the covalent binding of the 13C-labeled skin sensitizer 5-chloro-2-methylisothiazol-3-one to human serum albumin: identification of adducts by nuclear magnetic resonance, matrix-assisted laser desorption/ionization mass spectrometry, and nanoelectrospray tandem mass spectrometry Chem. Res. Toxicol. 2004b 17 1280 1288 Google Scholar Crossref Search ADS WorldCat Anraku M. Yamasaki K. Maruyama T. Kragh-Hansen U. Otagiri M. Effect of oxidative stress on the structure and function of human serum albumin Pharm. Res. 2001 18 632 639 Google Scholar Crossref Search ADS PubMed WorldCat Aptula A. O. Patlewicz G. Roberts D. W. Schultz T. W. Non-enzymatic glutathione reactivity and in vitro toxicity: A non-animal approach to skin sensitization Toxicol. In Vitro 2006 20 239 247 Google Scholar Crossref Search ADS PubMed WorldCat Ashby J. Basketter D. A. Paton D. Kimber I. Structure activity relationships in skin sensitization using the murine local lymph node assay Toxicology 1995 103 177 194 Google Scholar Crossref Search ADS PubMed WorldCat Basketter D. A. Gerberick G. F. Kimber I. Measurement of allergenic potency using the local lymph node assay Trends Pharmacol. Sci. 2001 22 264 265 Google Scholar Crossref Search ADS PubMed WorldCat Bertucci C. Domenici E. Reversible and covalent binding of drugs to human serum albumin: Methodological approaches and physiological relevance Curr. Med. Chem. 2002 9 1463 1481 Google Scholar Crossref Search ADS PubMed WorldCat Bhattacharya M. Jain N. Bhasne K. Kumari V. Mukhopadhyay S. pH-Induced conformational isomerization of bovine serum albumin studied by extrinsic and intrinsic protein fluorescence J. Fluoresc. 2011 21 1083 1090 Google Scholar Crossref Search ADS PubMed WorldCat Brander C. Mauri-Hellweg D. Bettens F. Rolli H. Goldman M. Pichler W. J. Heterogeneous T cell responses to beta-lactam-modified self-structures are observed in penicillin-allergic individuals J. Immunol. 1995 155 2670 2678 Google Scholar PubMed OpenURL Placeholder Text WorldCat Curry S. Mandelkow H. Brick P. Franks N. Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites Nat. Struct. Biol. 1998 5 827 835 Google Scholar Crossref Search ADS PubMed WorldCat Dennehy M. K. Richards K. A. Wernke G. R. Shyr Y. Liebler D. C. Cytosolic and nuclear protein targets of thiol-reactive electrophiles Chem. Res. Toxicol. 2006 19 20 29 Google Scholar Crossref Search ADS PubMed WorldCat Dietz L. Esser P. R. Schmucker S. S. Goette I. Richter A. Schnolzer M. Martin S. F. Thierse H. J. Tracking human contact allergens: From mass spectrometric identification of peptide-bound reactive small chemicals to chemical-specific naive human T-cell priming Toxicol. Sci. 2010 117 336 347 Google Scholar Crossref Search ADS PubMed WorldCat Elahi E. N. Wright Z. M. Hinselwood D. Hotchkiss S. A. Basketter D. A. Pease C. Protein binding and metabolism influence the relative skin sensitization potential of cinnamic compounds Chem. Res. Toxicol. 2004 17 301 310 Google Scholar Crossref Search ADS PubMed WorldCat Elbayed K. Berl V. Debeuckelaere C. Moussallieh F.-M. Piotto M. Namer I.-J. Lepoittevin J.-P. HR-MAS NMR spectroscopy of reconstructed human epidermis: Potential for the in situ investigation of the chemical interactions between skin allergens and nucleophilic amino acids Chem. Res. Toxicol. 2013 26 136 145 Google Scholar Crossref Search ADS PubMed WorldCat Franot C. Roberts D. W. Smith R. G. Basketter D. A. Benezra C. Lepoittevin J. P. Structure-activity relationships for contact allergenic potential of gamma,gamma-dimethyl-gamma-butyrolactone derivatives. 1. Synthesis and electrophilic reactivity studies of alpha-(omega-substituted-alkyl)-gamma,gamma-dimethyl-gamma-butyrolactones and correlation of skin sensitization potential and cross-sensitization patterns with structure Chem. Res. Toxicol. 1994 7 297 306 Google Scholar Crossref Search ADS PubMed WorldCat Gerberick G. F. Vassallo J. D. Bailey R. E. Chaney J. G. Morrall S. W. Lepoittevin J. P. Development of a peptide reactivity assay for screening contact allergens Toxicol. Sci. 2004 81 332 343 Google Scholar Crossref Search ADS PubMed WorldCat Gerberick G. F. Vassallo J. D. Foertsch L. M. Price B. B. Chaney J. G. Lepoittevin J. P. Quantification of chemical peptide reactivity for screening contact allergens: A classification tree model approach Toxicol. Sci. 2007 97 417 427 Google Scholar Crossref Search ADS PubMed WorldCat Geromanos S. J. Vissers J. P. Silva J. C. Dorschel C. A. Li G. Z. Gorenstein M. V. Bateman R. H. Langridge J. I. The detection, correlation, and comparison of peptide precursor and product ions from data independent LC-MS with data dependant LC-MS/MS Proteomics 2009 9 1683 1695 Google Scholar Crossref Search ADS PubMed WorldCat Griebenow K. Klibanov A. M. On protein denaturation in aqueous−organic mixtures but not in pure organic solvents J. Am. Chem. Soc. 1996 118 11695 11700 Google Scholar Crossref Search ADS WorldCat Jenkinson C. Jenkins R. E. Aleksic M. Pirmohamed M. Naisbitt D. J. Park B. K. Characterization of p-phenylenediamine-albumin binding sites and T-cell responses to hapten-modified protein J. Invest. Dermatol. 2010 130 732 742 Google Scholar Crossref Search ADS PubMed WorldCat Kato S. Seki T. Katsumura Y. Kobayashi T. Komatsu K. Fukushima S. Mechanism for 6-methylcoumarin photoallergenicity Toxicol. Appl. Pharmacol. 1985 81 295 301 Google Scholar Crossref Search ADS PubMed WorldCat Kesner L. Muntwyler E. Griffin G. E. Quaranta P. Hirs C. H. W. Determination of amino acids as DNP derivatives by automatic partition chromatography Methods In Enzymology 1967 11 NY Academic Press 94 108 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Landsteiner K. Jacobs J. Studies on the sensitization of animals with simple chemical compounds J. Exp. Med. 1935 61 643 656 Google Scholar Crossref Search ADS PubMed WorldCat Liebler D. C. Protein damage by reactive electrophiles: Targets and consequences Chem. Res. Toxicol. 2008 21 117 128 Google Scholar Crossref Search ADS PubMed WorldCat Loveless S. E. Ladics G. S. Gerberick G. F. Ryan C. A. Basketter D. A. Scholes E. W. House R. V. Hilton J. Dearman R. J. Kimber I. Further evaluation of the local lymph node assay in the final phase of an international collaborative trial Toxicology 1996 108 141 152 Google Scholar Crossref Search ADS PubMed WorldCat MacKay C. Davies M. Summerfield V. Maxwell G. From pathways to people: Applying the adverse outcome pathway (AOP) for skin sensitization to risk assessment Altex 2013 30 473 486 Google Scholar Crossref Search ADS PubMed WorldCat Maguire H. C. Jr Kaidbey K. Experimental photoallergic contact dermatitis: A mouse model J. Invest. Dermatol. 1982 79 147 152 Google Scholar Crossref Search ADS PubMed WorldCat Majeti V. A. Suskind R. R. Mechanism of cinnamaldehyde sensitization Contact Dermatitis 1977 3 16 18 Google Scholar Crossref Search ADS PubMed WorldCat Maxwell G. MacKay C. Cubberley R. Davies M. Gellatly N. Glavin S. Gouin T. Jacquoilleot S. Moore C. Pendlington R. et al. Applying the skin sensitisation adverse outcome pathway (AOP) to quantitative risk assessment Toxicol. In Vitro 2014 28 8 12 Google Scholar Crossref Search ADS PubMed WorldCat McFadden J. P. Puangpet P. Basketter D. A. Dearman R. J. Kimber I. Why does allergic contact dermatitis exist Br. J. Dermatol. 2013 168 692 699 Google Scholar Crossref Search ADS PubMed WorldCat Meschkat E. Barratt M. D. Lepoittevin J. P. Studies of the chemical selectivity of hapten, reactivity, and skin sensitization potency. 1. Synthesis and studies on the reactivity toward model nucleophiles of the (13)C-labeled skin sensitizers hex-1-ene- and hexane-1,3-sultones Chem. Res. Toxicol. 2001a 14 110 117 Google Scholar Crossref Search ADS WorldCat Meschkat E. Barratt M. D. Lepoittevin J. P. Studies of the chemical selectivity of hapten, reactivity, and skin sensitization potency. 2. nmr studies of the covalent binding of the (13)c-labeled skin sensitizers 2-[13C]- and 3-[13C]hex-1-ene- and 3-[13C]hexane-1,3-sultones to human serum albumin Chem. Res. Toxicol. 2001b 14 118 126 Google Scholar Crossref Search ADS WorldCat Michalski A. Cox J. Mann M. More than 100,000 detectable peptide species elute in single shotgun proteomics runs but the majority is inaccessible to data-dependent LC-MS/MS J. Proteome Res. 2011 10 1785 1793 Google Scholar Crossref Search ADS PubMed WorldCat Migdal C. Botton J. El Ali Z. Azoury M. E. Guldemann J. Gimenez-Arnau E. Lepoittevin J. P. Kerdine-Romer S. Pallardy M. Reactivity of chemical sensitizers toward amino acids in cellulo plays a role in the activation of the Nrf2-ARE pathway in human monocyte dendritic cells and the THP-1 cell line Toxicol. Sci. 2013 133 259 274 Google Scholar Crossref Search ADS PubMed WorldCat Natsch A. Gfeller H. LC-MS-based characterization of the peptide reactivity of chemicals to improve the in vitro prediction of the skin sensitization potential Toxicol. Sci. 2008 106 464 478 Google Scholar Crossref Search ADS PubMed WorldCat Natsch A. Gfeller H. Rothaupt M. Ellis G. Utility and limitations of a peptide reactivity assay to predict fragrance allergens in vitro Toxicol. In Vitro 2007 21 1220 1226 Google Scholar Crossref Search ADS PubMed WorldCat OECD The adverse outcome pathway for skin sensitisation initiated by covalent binding to proteins. Part 2: Use of the AOP to develop chemical categories and integrated assessment and testing approaches 2012 168 1 46 OECD Environment, Health and safety Publications Series on Testing and Assessment OpenURL Placeholder Text WorldCat Google Scholar Parkinson E. Skipp P. Aleksic M. Garrow A. Dadd T. Hughes M. Clough G. O'Connor C. D. Proteomic analysis of the human skin proteome after in vivo treatment with sodium dodecyl sulphate PLoS ONE 2014 9 e97772 Google Scholar Crossref Search ADS PubMed WorldCat Roberts D. W. Aptula A. O. Patlewicz G. Electrophilic chemistry related to skin sensitization. Reaction mechanistic applicability domain classification for a published data set of 106 chemicals tested in the mouse local lymph node assay Chem. Res. Toxicol. 2007 20 44 60 Google Scholar Crossref Search ADS PubMed WorldCat Santra M. K. Banerjee A. Rahaman O. Panda D. Unfolding pathways of human serum albumin: Evidence for sequential unfolding and folding of its three domains Int. J. Biol. Macromol. 2005 37 200 204 Google Scholar Crossref Search ADS PubMed WorldCat Schirle M. Heurtier M. A. Kuster B. Profiling core proteomes of human cell lines by one-dimensional PAGE and liquid chromatography-tandem mass spectrometry Mol. Cell. Proteomics 2003 2 1297 1305 Google Scholar Crossref Search ADS PubMed WorldCat Shevchenko A. Jensen O. N. Podtelejnikov A. V. Sagliocco F. Wilm M. Vorm O. Mortensen P. Boucherie H. Mann M. Linking genome and proteome by mass spectrometry: Large-scale identification of yeast proteins from two dimensional gels Proc. Natl Acad. Sci. U.S.A. 1996 93 14440 14445 Google Scholar Crossref Search ADS PubMed WorldCat Silva J. C. Gorenstein M. V. Li G. Z. Vissers J. P. Geromanos S. J. Absolute quantification of proteins by LCMSE: A virtue of parallel MS acquisition Mol. Cell. Proteomics 2006 5 144 156 Google Scholar Crossref Search ADS PubMed WorldCat Simonsson C. Andersson S. I. Stenfeldt A.-L. Bergstrom J. Bauer B. Jonsson C. A. Ericson M. B. Broo K. S. Caged fluorescent haptens reveal the generation of cryptic epitopes in allergic contact dermatitis J. Invest. Dermatol. 2011 131 1486 1493 Google Scholar Crossref Search ADS PubMed WorldCat Thierse H. J. Moulon C. Allespach Y. Zimmermann B. Doetze A. Kuppig S. Wild D. Herberg F. Weltzien H. U. Metal-protein complex-mediated transport and delivery of Ni2+ to TCR/MHC contact sites in nickel-specific human T cell activation J. Immunol. 2004 172 1926 1934 Google Scholar Crossref Search ADS PubMed WorldCat Tzouros M. Pahler A. A targeted proteomics approach to the identification of peptides modified by reactive metabolites Chem. Res. Toxicol. 2009 22 853 862 Google Scholar Crossref Search ADS PubMed WorldCat Vila A. Tallman K. A. Jacobs A. T. Liebler D. C. Porter N. A. Marnett L. J. Identification of protein targets of 4-hydroxynonenal using click chemistry for ex vivo biotinylation of azido and alkynyl derivatives Chem. Res. Toxicol. 2008 21 432 444 Google Scholar Crossref Search ADS PubMed WorldCat Wetzel R. Becker M. Behlke J. Billwitz H. BöHm S. Ebert B. Hamann H. Krumbiegel J. Lassmann G. Temperature behaviour of human serum albumin Eur. J. Biochem. 1980 104 469 478 Google Scholar Crossref Search ADS PubMed WorldCat © The Author 2014. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please email: journals.permissions@oup.com
TI - Stable Isotope Labeling Method for the Investigation of Protein Haptenation by Electrophilic Skin Sensitizers
JF - Toxicological Sciences
DO - 10.1093/toxsci/kfu168
DA - 2014-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/stable-isotope-labeling-method-for-the-investigation-of-protein-hcZhO1L0K2
SP - 239
EP - 249
VL - 142
IS - 1
DP - DeepDyve
ER -