Elsevier

Toxicology in Vitro

Volume 21, Issue 4, June 2007, Pages 723-733
Toxicology in Vitro

Investigating protein haptenation mechanisms of skin sensitisers using human serum albumin as a model protein

https://doi.org/10.1016/j.tiv.2007.01.008Get rights and content

Abstract

Covalent modification of skin proteins by electrophiles is a key event in the induction of skin sensitisation but not skin irritation although the exact nature of the binding mechanisms has not been determined empirically for the vast majority of sensitisers. It is also unknown whether immunologically relevant protein targets exist in the skin contributing to effecting skin sensitisation. To determine the haptenation mechanism(s) and spectra of amino acid reactivity in an intact protein for two sensitisers expected to react by different mechanisms, human serum albumin (HSA) was chosen as a model protein. The aim of this work was also to verify for selected non-sensitisers and irritants that no protein haptenation occurs even under forcing conditions. HSA was incubated with chemicals and the resulting complexes were digested with trypsin and analysed deploying matrix-assisted laser desorption/ionization mass spectrometry, reverse phase high performance liquid chromatography and nano-electrospray tandem mass spectrometry. The data confirmed that different residues (lysine, cysteine, histidine and tyrosine) are covalently modified in a highly selective and differential manner by the sensitisers 2,4-dinitro-1-chlorobenzene and phenyl salicylate. Additionally, non-sensitisers 2,4-dichloro-1-nitrobenzene, butyl paraben and benzaldehyde and irritants benzalkonium chloride and sodium dodecyl sulphate did not covalently modify HSA under any conditions. The data indicate that covalent haptenation is a prerequisite of skin sensitisation but not irritation. The data also suggest that protein modifications are targeted to certain amino acids residing in chemical microenvironments conducive to reactivity within an intact protein. Deriving such information is relevant to our understanding of antigen formation in the immunobiology of skin sensitisation and in the development of in vitro protein haptenation assays.

Introduction

Skin exposure to low molecular weight chemicals (typically <500 Da (Bos and Meinardi, 2000)) can result in specific immunochemical events occurring at the molecular and cellular level that can manifest themselves as allergic contact dermatitis (ACD). ACD is a delayed type hypersensitivity reaction, the first phase of which is the induction of skin sensitisation, a chemical-specific priming of the immune system (Coombs and Gell, 1975, Girolomoni et al., 2004, Rustemeyer et al., 2006), which is manifested by the occurrence of erythema and oedema at the site of secondary skin contact with a small chemical allergen in the elicitation phase.

The chemistry of skin sensitisation has not been studied extensively and much of our knowledge is theoretical. The key events in the process of skin sensitisation are often described based on the Coombs and Gell Coombs and Gell (1975) classification relating to type IV delayed type hypersensitivity and are as follows. Small electrophilic chemical sensitisers (haptens) are not large enough to stimulate an immune response and must be complexed with skin protein(s) producing covalent adducts (Landsteiner and Jacobs, 1935, Dupuis and Benezra, 1982, Lepoittevin et al., 1998, Smith and Hotchkiss, 2001). Molecules that are not inherently reactive towards proteins can also act as sensitisers, once activated by skin metabolism or air oxidation into protein reactive species (Karlberg et al., 1992, Bergh et al., 1998, Smith and Hotchkiss, 2001, Skold et al., 2004, etc.). Covalently modified skin protein(s) are internalised and processed by the Langerhans cells. Langerhans cells mature and migrate to the draining lymph node where they present haptenated peptides to naïve T cells in association with major histocompatibility complex (MHC). Presented antigenic peptide interacts with the T cell receptor and, assisted by a number of co-stimulatory factors expressed by both cell types, T cells recognise the haptenated peptide as foreign. The end result is clonal expansion of antigen-specific T cells. When an individual has a population of antigen-specific T cells in circulation, they are said to be sensitised to the specific chemical that was first encountered. The clinical condition of ACD becomes manifest following subsequent skin contact with the same or cross-reactive chemicals that lead to the same type of haptenated antigens.

The skin sensitisation potential of chemicals is currently assessed using the murine local lymph node assay (LLNA), which enables hazard characterisation and allows an estimation of sensitisation potency of the tested chemicals (Basketter et al., 2001). However, as an in vivo test, the LLNA will become unavailable with the implementation of the EU ban on in vivo testing of cosmetic and toiletry ingredients (EU, 2003). As one of the key molecular events in the skin sensitisation process, protein–hapten binding data could be used in novel integrated approaches for the prediction of skin sensitisation potential of chemicals (Jowsey et al., 2006). It is therefore essential to understand the reactions of sensitising chemicals with (skin) proteins to be able to discriminate between sensitisers and non-sensitisers. Generating reaction mechanism and specificity based knowledge will help in development of novel alternative in vitro test methods to generate useful data for prediction of skin sensitisation potential.

Previous investigations into the chemical mechanisms of how skin sensitisers bind to protein nucleophiles have been performed using isolated model nucleophiles such as butylamine, N-acetyl-cysteine, etc. (Franot et al., 1994, Meschkat et al., 2001a, Alvarez-Sanchez et al., 2003). However, such studies do not reveal the total spectrum of reactivity or any preferential binding that may occur, when incubating a chemical with a large intact protein containing all potential nucleophiles simultaneously. In choosing a potential protein for this work, it needed to be well characterised and relevant to the skin. The protein profile of human skin has not yet been determined, but studies using keratinocyte cultures have shown that the number of proteins expressed is substantial (Celis et al., 1998). Human keratinocyte databases resulting from these studies (http://proteomics2.cancer.dk/cgi-bin/CelisWeb.exe?MsetList.htm) list around 850 known proteins of almost 2500 proteins detected. It has not yet been established whether certain proteins in the skin are targeted by skin sensitisers. Therefore, to investigate protein binding mechanisms of sensitising chemicals we chose to work with human serum albumin (HSA) as a representative protein abundant within the skin.

HSA (MW 66 472 Da) is found mainly in plasma (∼50%) where it maintains the pH and osmotic pressure and has a major role in transporting a wide range of molecules such as metals, fatty acids, amino acids, metabolites and many drugs. A large proportion of extravascular albumin is found in the skin (41%) (Peters, 1996). Numerous studies have been conducted using HSA and other albumins as model proteins (e.g. Bertucci et al., 1995, Tracey and Shuker, 1997, Yasuzawa and Tomer, 1997, Moser et al., 2000, Meschkat et al., 2001b).

To investigate protein–hapten binding hypotheses further, we explored the spectrum of reactivity of known sensitisers 2,4-dinitro-1-chlorobenzene (DNCB) and phenyl salicylate (PS) (Fig. 1, 1 and 5, respectively). Theoretical reaction mechanisms for DNCB and PS (Fig. 1, 8 and 9, respectively) suggest that only a single reaction mechanism for each chemical yielding one type of adduct is likely. Non-sensitisers 2,4-dichloro-1-nitrobenzene (DCNB), benzaldehyde (BZ) and butyl paraben (BP) (Fig. 1, 24, respectively) and irritants sodium dodecyl sulphate (SDS) and benzalkonium chloride (BC) (Fig. 1, 6 and 7, respectively) were also incubated with HSA and were not expected to bind. However, non-sensitiser BZ could potentially react with nucleophiles via Schiff base formation (Fig. 1, 10). Irritant responses are thought to arise from the direct toxic effect of the irritant chemical on the cells exposed to it, rather than the ability of chemical to covalently modify self protein(s). Therefore, we investigated whether the covalent modifications are specific and relevant for sensitisation, as this is a prerequisite for an alternative assay for sensitisation.

Section snippets

Sample preparation

All test chemicals were supplied by Sigma as >95% pure and solubilised in 3:2 dimethylsulphoxide:water (v/v). Samples were prepared using solutions of test chemicals and solution of HSA (>99%, fatty acid free, Sigma) at a 1:100 and 1:1000 molar ratio of protein:chemical in 50 mM ammonium acetate (as HSA has 125 nucleophilic side chains the molar ratios of nucleophile:electrophile are effectively 1:0.8 and 1:8, respectively). Incubation buffers were maintained at pH 5.5, 6.8 or 7.4 (reflecting

HSA peptide mapping and assessment of experimental variability

On average, more than 90% of the native HSA sequence was observed in any one MALDI-ToF spectrum of the control samples (Fig. 2a and b). Relative signal intensities changed depending on the position of the laser beam on the sample-matrix mixture as well as laser intensity at the time of data acquisition, thus only semiquantitative information can be derived when MALDI-MS spectra are compared. Although observing reduced signal intensity could potentially mean that the relevant peptide is modified

Discussion

In this study, we have revealed the promiscuous nature of binding of DNCB to a broad spectrum of nucleophiles in the intact protein HSA, and as expected this binding was via an SnAr mechanism yielding a DNP adduct with resulting mass increase of 166 Da. The binding observed was selective for nucleophiles that are presumably in an environment more conducive to reactivity than others or perhaps more accessible in the 3D globular fold of the protein. The binding of PS showed a different spectrum of

Acknowledgements

The authors thank the BBSRC and Unilever Research for the BBSRC CASE studentship for M.A. and BBSRC and Wellcome Trust for funding. A.D. is a BBSRC Professorial Research Fellow. Additionally, the authors thank Dr. Diana Drennan, Unilever R&D, Trumbull, USA, for help with visualisation of protein modifications.

References (45)

  • J. Moser et al.

    Photobinding of ketoprofen in vitro and ex vivo

    J. Photochem. Photobiol. B

    (2000)
  • D. Parker et al.

    A comparison of the conjugation of DNTB and other dinitrobenzenes with free protein radicals and their ability to sensitize or tolerize

    J. Invest. Dermatol.

    (1983)
  • J.E. Walker

    Lysine residue 199 of human serum albumin is modified by acetylsalicyclic acid

    FEBS Lett.

    (1976)
  • R. Alvarez-Sanchez et al.

    Studies of chemical selectivity of hapten, reactivity, and skin sensitization potency. 3. Synthesis and studies on the reactivity toward model nucleophiles of the 13C-labelled skin sensitizers, 5-chloro-2-methylisothiazol-3-one (MCI) and 2-methylisothiazol-3-one (MI)

    Chem. Res. Toxicol.

    (2003)
  • R. Alvarez-Sanchez et al.

    Effect of glutathione on the covalent binding of the 13C-labelled 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.

    (2004)
  • D.A. Basketter et al.

    Dichloronitrobenzene: a reappraisal of its skin sensitization potential

    Contact Dermatitis

    (1996)
  • M. Bergh et al.

    Formation of formaldehyde and peroxides by air oxidation of high purity polyoxyethylene surfactants

    Contact Dermatitis

    (1998)
  • J.D. Bos et al.

    The 500 Dalton rule for the skin penetration of chemical compounds and drugs

    Exp. Dermatol.

    (2000)
  • M.M. Carr et al.

    Early cellular reactions induced by dinitrochlorobenzene in sensitized human skin

    Br. J. Dermatol.

    (1984)
  • R.R.A. Coombs et al.

    Classification of allergic reactions for clinical hypersensitivity and disease

  • G. Dupuis et al.

    Contact Dermatitis to Simple Chemicals: A Molecular Approach

    (1982)
  • E.N. Elahi et al.

    Protein binding and metabolism influence the relative skin sensitization potential of cinnamic compounds

    Chem. Res. Toxicol.

    (2004)
  • Cited by (60)

    • High-throughput screening (HTS)-based spectrophotometric direct peptide reactivity assay (Spectro-DPRA) to predict human skin sensitization potential

      2019, Toxicology Letters
      Citation Excerpt :

      In particular, the evaluation of the reactivity between chemicals and proteins/peptides serves as an attractive alternative to animal testing to determine skin sensitization potential of test compounds (Divkovic et al., 2005; Gerberick et al., 2008). At present, the measurement of reactivity (adduct formation) between a chemical substance and protein/peptide largely relies on two analytical methods, namely, high-performance liquid chromatography (HPLC) with ultraviolet (UV) detection (Gerberick et al., 2004, 2007) and liquid chromatography-mass spectrometry (LC/MS) (Ahlfors et al., 2005; Aleksic et al., 2007, 2009; Alvarez-Sanchez et al., 2004; Natsch and Gfeller, 2008). However, HPLC- or LC/MS-based methods have several disadvantages, including the requirement of relatively expensive equipments for analysis, long sample measurement procedures, and skilled operators.

    • Correlating the structure and reactivity of a contact allergen, DNCB, and its analogs to sensitization potential

      2019, Bioorganic and Medicinal Chemistry
      Citation Excerpt :

      After 20 h of incubation, the cells were harvested and the supernatant was collected for analysis. Conjugation of target protein was performed according to previously published protocol.39–41 Bovine serum albumin (BSA, Millipore Sigma) and the selected compounds were dissolved in a mixture composed of 17 parts 1X PBS and 3 parts 100% acetonitrile (HPLC grade) and incubated in 37 °C for 7 days.

    • Contact Hypersensitivity

      2017, Comprehensive Toxicology: Third Edition
    View all citing articles on Scopus
    View full text