An automated method for the quantification of immunostained human Langerhans cells

https://doi.org/10.1016/S0022-1759(00)00328-8Get rights and content

Abstract

Allergic contact dermatitis is a frequent and increasing health problem. For ethical reasons, the current animal tests used to screen for contact sensitizers should be replaced by in vitro alternatives. Contact sensitizers have been shown to accelerate Langerhans cell (LC) migration from human organotypic skin explant cultures (hOSECs) more rapidly than non-sensitizers and it has been proposed that the hOSEC model could be used to screen for sensitizers. However, chemically induced decreases in epidermal LC numbers need to be accurately quantified if the alterations in epidermal LC numbers are to form the basis of an alternative system for screening contact sensitizers in vitro. As manual counting of LCs is labour intensive and subject to intra- and inter-personal variation we developed an image analysis routine, using the Leica QWin image analysis software, to quantify LCs in situ using immunohistochemically stained skin sections. LCs can be identified using antibodies against the membrane molecule CD1a or the Lag antibody, which recognises cytoplasmic Birbeck granules. Quantification of epidermal LC number using the image analysis software had a much lower inter-person variation than when the same specimens were counted manually, using both the anti-Lag and CD1a antibodies. The software-aided quantification of epidermal LCs provides an accurate method for measuring chemically-induced changes in LC numbers.

Introduction

Allergic contact dermatitis is a frequent health problem. Contact allergens are currently screened using animal models, such as the guinea pig maximisation test (GPMT) and the murine local lymph node assay (LLNA). The assessment of the sensitisation potential of a single chemical requires 24 to 32 guinea pigs or 16 to 30 mice. The accuracy of both the GPMT and the LLNA models for predicting human contact sensitisers is about 70% (Anonymous, 1999). Differences in the response of the immune system and skin morphology could account for part of the low efficiencies (Bouclier et al., 1990).

Immature dendritic cells (DCs), such as Langerhans cells (LCs) in the epidermis, take up antigen in the peripheral tissue (Shelley and Juchlin, 1976). After activation, e.g., induced by contact allergens, LCs migrate to the draining lymph node and mature (Aiba et al., 1997, Drexhage et al., 1979, Macatonia et al., 1986, Silberberg et al., 1976). Mature LCs or DCs stimulate the development of hapten-specific naive T cells leading to antigen-specific sensitisation (Inaba et al., 1986, Soeberg et al., 1978, Streilein, 1989). The subsequent application of a contact allergen on the skin elicits an allergic contact dermatitis (Roitt et al., 1998).

Migration of epidermal LCs can be studied in human organotypic skin explant cultures (hOSECs), where LCs spontaneously migrate out through lymphatic vessels (Czernielewski et al., 1984, Lukas et al., 1996, Rambukkana et al., 1995). The topical exposure of hOSEC to contact sensitisers accelerates LC migration out of the epidermis, relative to spontaneous migration and the migration induced by control chemicals. This acceleration of epidermal LC migration induced by a contact sensitizer may be used as a screening system for contact allergens (Pistoor et al., 1996, Rambukkana et al., 1996). Manual counting of epidermal LCs is labour intensive and subject to intra- and inter-personal variation. For these reasons we developed an image analysis routine using Leica QWin image analysis software which can be used to quantify LCs in immunohistochemically stained skin sections in situ.

Section snippets

Human organotypic skin explant cultures (human OSECs, hOSECs)

Dulbecco’s phosphate-buffered saline (DPBS) (BioWhittaker, Verviers, Belgium), mineral oil, nickel sulfate, potassium dichromate and sodium dodecyl sulfate (SDS) (all Sigma–Aldrich, Zwijndrecht, The Netherlands) were preheated to 37°C, prior to application onto the skin. Human breast skin was obtained as a waste product of cosmetic surgery. Sterile biopsies were cut (approx. 0.25 cm2) and these were incubated dermal-side down in Dulbecco’s Modified Eagles Medium with ultraglutamine 1, with 4.5

Visual examination of eLC stainings

LCs were stained in cryostat sections using MHC-II, CD1a, or Lag antibodies (Fig. 2A–C). MHC-II staining of human LCs was not specific as in some experiments all keratinocytes in the epidermis were MHC-II positive (data not shown). For the studies reported in this article, skin LCs were defined as Lag+ or CD1a+. In practice, all epidermal and dermal CD1a+ cells were Lag+ and vice versa (data not shown). Lag stains the Birbeck granules, which are present in the LC body, and positive cells appear

Detection of LCs

Epidermal LCs can be visualised using a number of unique markers. Adequate and accurate counting of LCs requires that these markers are stable and uniquely expressed on LCs. We stained LCs in cryostat sections with three antibodies, anti-MHC-II, CD1a and Lag (Fig. 2A–C). LCs could be visualised using CD1a or Lag staining and all epidermal and dermal LCs double stained for both CD1a and Lag markers (data not shown). Our results are in agreement with in vivo data showing that LCs emigrating from

Acknowledgements

This study was subsidised by a grant from the Dutch Platform Alternatives for Animal Experiments to G.R.E. and P.K.D. for the support of J.J.L.J. and C.L. (grant No. 96-32). P.K.D. also acknowledges grant support from the Dr. Hadwen Trust Research in Humanities, UK.

References (27)

  • M. Bouclier et al.

    Experimental models in skin pharmacology

    Pharmacol. Rev.

    (1990)
  • C.U. Brand et al.

    Studies on human skin lymph containing Langerhans cells from sodium lauryl sulfate contact dermatitis

    J. Invest. Dermatol.

    (1992)
  • J. Czernielewski et al.

    Human Langerhans cells in epidermal cell culture, in vitro skin explants and skin grafts onto “nude” mice

    Arch. Dermatol. Res.

    (1984)
  • Cited by (22)

    View all citing articles on Scopus
    View full text