Elsevier

The Lancet

Volume 362, Issue 9378, 12 July 2003, Pages 118-123
The Lancet

Mechanisms of Disease
Autoantibodies to extracellular matrix protein 1 in lichen sclerosus

https://doi.org/10.1016/S0140-6736(03)13863-9Get rights and content

Summary

Background

Lichen sclerosus is a common acquired inflammatory disorder of skin and mucous membranes. The aetiology is unknown, although HLA-subtype susceptibility and high rates of other autoimmune disorders suggest that autoantibodies to specific mucocutaneous antigens are involved. The clinicopathological similarities between lichen sclerosus and lipoid proteinosis, which results from mutations in extracellular matrix protein 1 (ECM1), suggest this protein as an autoantigen.

Methods

We analysed serum autoantibody profiles in 171 individuals (86 with lichen sclerosus, 85 healthy controls) by immunoblotting of extracts from normal human skin and lipoid proteinosis skin (lacking ECM1). We generated a full-length glutathione-S-transferase fusion protein for ECM1 to confirm specific immunoreactivity. We affinity-purified serum from patients with lichen sclerosus and did indirect immunofluorescence microscopy on normal skin with or without preabsorption with recombinant ECM1.

Findings

By immunoblotting, IgG autoantibodies were found in 20 (67% [95% CI 45–84]) of 30 lichen sclerosus serum samples. The highest titre was 1 in 20. The bands were not detected in ECM1–deficient substrate. These samples, and those from 56 other patients with lichen sclerosus, showed immunoreactivity to the recombinant ECM1 protein (64 of 86 positive; 74% [65–84]). Only six (7% [2–13]) of 85 control serum samples were positive. Affinity-purified IgG from serum of patients with lichen sclerosus labelled skin similarly to a polyclonal antibody to ECM1. The positive staining was blocked by preabsorption with excess recombinant ECM1 protein.

Interpretation

These findings provide evidence for a specific humoral immune response to ECM1 in lichen sclerosus and offer insight into disease diagnosis, monitoring, and approaches to treatment.

Introduction

Lichen sclerosus is a chronic inflammatory disorder of unknown cause.1, 2, 3 Typically, it consists of pale indurated papules and plaques that most commonly affect the genitalia. Clinical and epidemiological studies suggest that the disorder is under-reported and may have a prevalence of more than one in 1000.2, 4 Women and girls are more commonly affected than men, with the ratio varying from ten to six to one.1, 2 Lichen sclerosus can occur at any age, but in the female population there is a bimodal peak in prepubertal children and after the menopause, and in the male population the disorder is most common at 30–50 years of age.2 Symptoms can include intractable itching and soreness, and the chronic skin and mucous-membrane inflammation can lead to scarring. Some longstanding cases are complicated by malignant disease, most commonly squamous-cell carcinoma.2 Extragenital involvement generally occurs on the head and neck, but it has been reported in the scalp, palms and soles, around the eye, the oral mucosa, the tongue, and around stoma sites.1, 2

Histology of affected skin in lichen sclerosus shows hydropic degeneration of basal keratinocytes and a homogeneous appearance of the collagen in the upper dermis. In some cases there is inflammation and disruption of the basement membrane as well as inflammatory changes in all zones of the skin.3 Trauma and infection have been implicated in the aetiology of lichen sclerosus, but the precise pathogenesis is not known. Nevertheless, accumulating evidence suggests a genetic susceptibility and autoimmune basis to the disorder.1, 3, 5 For example, familial cases of lichen sclerosus have been reported in identical and non-identical twins, siblings, and mothers and daughters.6, 7 Furthermore, some reports have identified a disease association with the class II HLA antigen DQ7.8, 9 This immunogenetic association has also been found in another mucocutaneous scarring disorder, mucous-membrane pemphigoid.10 Finally, several other studies have shown an increased rate of autoimmune disease and autoantibodies in individuals with lichen sclerosus: associated disorders include thyroid disease, pernicious anaemia, diabetes mellitus, alopecia areata, vitiligo, and mucous-membrane pemphigoid.1, 5, 11

Although little progress has been made in unravelling a specific disease mechanism in lichen sclerosus, there have been substantial recent advances in characterising the immunopathological basis of several acquired skin scarring disorders associated with basement-membrane pathology, such as mucous-membrane pemphigoid. Specifically, autoantibodies to the epithelial-basement-membrane components type XVII collagen (also known as the 180 kDa bullous pemphigoid antigen, BP180 or BPAG2) and laminin 5 have been identified.12, 13 Other studies have shown that inherited mutations in these proteins also result in mucocutaneous fragility and scarring, notably in a group of diseases known as junctional epidermolysis bullosa.14, 15 Several other structural proteins in skin and mucous membranes (eg, type VII collagen, a6(34 integrin, and plectin) have now been identified that serve as targets for either autoantibodies (acquired disease) or gene mutations (inherited disease). Therefore, one possibility is that part of the unexplained disease mechanism in lichen sclerosus involves development of autoantibodies against an as yet unidentified target antigen that is present in skin and certain mucous membranes. One putative candidate antigen is the glycoprotein extracellular matrix protein 1 (ECM1). This idea is based on the recent identification of pathogenetic mutations in the ECM1 gene in the autosomal recessive genetic disorder lipoid proteinosis,16 also known as hyalinosis cutis at mucosae (OMIM 247100). Notably, molecular studies established that lipoid proteinosis results from loss-of-function mutations in ECM1.16, 17 Some individuals with lipoid proteinosis have clinicopathological features that overlap with those of lichen sclerosus. These include trauma-induced inflammation of skin and to a lesser extent mucous membranes and, most importantly, very similar hyaline (glassy) changes in the dermis on skin-biopsy samples (figure 1). Other studies have highlighted a role for ECM1 protein in epidermal differentiation and binding of dermal matrix proteins (eg, to the major heparin sulphate proteoglycan, perlecan), as well as stimulation of vascular endothelial cells in tumour stroma.18, 19, 20 Since the features of lichen sclerosus include epidermal and epithelial atrophy and hyperkeratosis, dermal fibrosis, purpura, and an increased risk of malignant disease, the possibility that autoantibodies disrupt the function of ECM1 in lichen sclerosus is plausible. Therefore, we investigated whether lichen sclerosus is associated with the presence of autoantibodies to ECM1.

Section snippets

Patients

We assessed serum samples from 86 patients with lichen sclerosus (all women over the age of 40 years) and 85 healthy controls matched for age and sex. All the patients with lichen sclerosus had genital disease, but seven also had extragenital involvement. The diagnosis of lichen sclerosus was confirmed by skin biopsy in more than 70% of patients. Archived serum samples from 107 further individuals with other autoimmune basement-membrane or sclerosing diseases (22 systemic sclerosis, 45 systemic

Results

Initial immunoblotting with serum samples from 30 patients with lichen sclerosus showed that discrete bands (roughly 61 kDa, 58 kDa, and 48 kDa) were detectable in 20 (67% [95% CI 45–84]) of 30 samples. These bands were observed with extracts from normal human epidermis, dermis, and cultured keratinocytes (figure 2). Immunoreactivity was higher to the epidermal or keratinocyte substrates than to the dermal substrate, which suggests relative differences in ECM1 expression in the different skin

Discussion

Our findings provide evidence for the presence of circulating autoantibodies to a specific skin protein, ECM1, in most patients with lichen sclerosus. The study has not shown that these antibodies are pathogenetic, but the findings offer further evidence for humoral autoimmunity in the aetiology of this chronic inflammatory disorder.

The ECM1 antibodies are present at low concentrations in the serum, in keeping with previous studies that have not found bound antibodies (apart from fibrin) in

GLOSSARY

affinity chromatography/purification
Purification system to separate proteins with a particular antigenic determinant from a mixture of molecules.
bradford protein assay
A colorimetric assay using dye, which binds to aminoacid residues, that is commonly used to calculate protein concentration.
hydropic degeneration
Disruption of the integrity of keratinocytes within the basal layer of the epidermis.
kozak translation initation sequence
A specific genomic DNA sequence immediately upstream from the

References (30)

  • D Zillikens

    Acquired skin diseases of hemidesmosomes

    J Dermatol Sci

    (1999)
  • GL Tasker et al.

    Lichen sclerosus

    Clin Exp Dermatol

    (2003)
  • HJ Wallace

    Lichen sclerosus et atrophicus

    Trans St John's Hosp Dermatol Soc

    (1971)
  • RH Meyrick Thomas et al.

    Lichen sclerosus et atrophicus and autoimmunity: a study of 350 women

    Br J Dermatol

    (1998)
  • RH Meyrick Thomas et al.

    The development of lichen sclerosus et atrophicus in monozygotic twin girls

    Br J Dermatol

    (1986)
  • Cited by (255)

    • Targeted Therapies in Autoimmune Skin Diseases

      2022, Journal of Investigative Dermatology
    View all citing articles on Scopus
    View full text