Why Lesions Occur at Different Sites

By Grant J. Anhalt, M.D.
Johns Hopkins Dermatology

I will attempt to clarify what we know about the antibody response in various forms of pemphigus and how the distribution of the targeted antigens affects the location of lesions. The synthesis of this work has been proposed by Dr. John Stanley, with key published advances from Dr. Masa Amagai and Mai Mahoney, Ph.D., P. Koch and others. John Stanley refers to his concept as the “desmoglein compensation hypothesis”. The key to this hypothesis is the desmogleins (pemphigus antigens) are key adhesion molecules that keep cells attached to each other. In some areas of the body, there are two desmogleins present, and both have to be damaged to cause cell detachment – in some areas only one desmoglein may be present at some level in the skin or mucous membrane, and there only one desmoglein has to be damaged to cause cell detachment.

The following summary is extracted from a recent editorial from the Journal of the American Academy of Dermatology (1). Form of pemphigus Antigens targeted by autoantibodies Pemphigus foliaceus Desmoglein 1 Pemphigus vulgaris (only oral lesions) Desmoglein 3 Pemphigus vulgaris (oral and skin lesions) Desmoglein 3 and Desmoglein 1 Paraneoplastic pemphigus Desmoglein 3, Desmoglein 1 and Plakin proteins.

The simplest situation is pemphigus foliaceus, where blistering occurs in the most superficial layer of the epidermis. These patients have antibodies that are directed against desmoglein 1. This desmoglein is expressed mostly in the upper levels of the epidermis, and antibodies binding to it cause cell detachment (blistering) only where it is primarily expressed. Desmoglein 1 is present in oral epithelium, but oral lesions “never” occur because there is co-expression of desmoglein 3 in the superficial oral epithelium, and the desmoglein 3 is able to keep the cells from detaching, even if desmoglein 1 is knocked out by the antibody (2 ). Also, in skin, desmoglein 3 is expressed in lower levels of the epidermis, so that knocking out desmoglein 1 doesn’t cause blistering there, just in the upper levels of the skin where desmoglein 1 is the only key adhesion molecule present. In pemphigus vulgaris, antibodies impair the adhesive function of desmoglein 3 and cause acantholysis in the oral cavity, where this molecule seems to be of primary importance in maintaining cell attachment. The hypothesis that desmoglein 3 has such unique contributions to adhesion primarily in oral epithelium was is supported by new data. Recently, a “knockout” mouse was produced, where desmoglein 3 was eliminated by genetic manipulation (3). These mice lacked desmoglein 3 expression in skin and mucous membranes, but acantholysis occurred only in the mouth, and in a limited fashion in areas of trauma such as the snout and paws. The skin was not affected.

As pemphigus vulgaris progresses, many, but not all patients will develop cutaneous disease. This seems to be caused by a “spread” of the autoantibody response, and development of antibodies against both desmoglein 3 and desmoglein 1. If both these desmogleins are inhibited by autoantibody, not just oral, but also cutaneous blistering occurs (4, 5). This may represent an example of what has been called “epitope spreading” (6). This is a phenomenon seen in autoimmunity, where early in the disease course, patients make antibodies against one protein, but as the disease evolves, they then make additional antibodies against molecules that are similar in structure, or even unlike each other in structure, but are physically closely associated in a tissue.

Patients with pemphigus foliaceus generally do not develop epitope spreading. If they did, they would start to make antibodies against desmoglein 3 (as well as 1) and would then have pemphigus vulgaris. This has been observed, but is actually a very rare event. Also, one can not have both pemphigus vulgaris and foliaceus, because pemphigus foliaceus is defined by the absence of antibodies against desmoglein 3.

References

1. Anhalt GJ. Making sense of antigens and antibodies in pemphigus (Editorial). J Amer Acad Dermatol, 40;763-766,1999.

2. Shirakata Y, Amagai M, Hanakawa Y, et al. Lack of mucosal involvement in pemphigus foliaceus may be due to low expression of desmoglein 1.J Invest Dermatol 1998;110:76-78,

3. Koch PJ, Mahoney MG, Ishikawa H, et al. Targeted disruption of the pemphigus vulgaris antigen (desmoglein 3) gene in mice causes loss of keratinocyte cell adhesion with a phenotype similar to pemphigus vulgaris. J Cell Biol 1997:137;1091-1102.

4. Amagai M, Koch PJ, Nishikawa T, et al. Pemphigus vulgaris antigen is localized in the lower epidermis, the site of blister formation in patients. J Invest Dermatol 1996; 106:351-355.

5. Ding X, Aoki V, Mascaro JM, et al. Mucosal and mucocutaneous (generalized) pemphigus vulgaris show distinct autoantibody profiles. J Invest Dermatol 1997;109:592-596. 6. Craft J, Fatenejad S. Self antigens and epitope spreading in systemic autoimmunity. Arthritis Rheumatism 1997:40;1374-1382.