Department of Veterinary Internal Medicine, Tokyo University of Agriculture & Technology
Desmosomes bind epithelial cells to the adjacent cells via their transmembrane structures for the integration of the epidermis. Desmosomes are composed of transmembraneous proteins including desmogleins (Dsgs), desmocollins, and desmoplakins, and intracellular supporting proteins, such as plakoglobin. The transmembrane proteins are responsible for the onset of canine pemphigus foliaceus (PF) and pemphigus vulgaris (PV). Clinical signs of pemphigus may appear when the antigen-antibody reaction occurs at the sites of desmosomes, but the mechanisms underlying the digestion of the desmosomes after the antigen-antibody reaction have not been explored so far. Canine Dsg1 and Dsg3 and desmocollins 1 and 2 have been cloned and sequenced and Dsg4 to 6 have been in mice.
Desmoglein 1 (Dsg1)
Desmoglein 1 as target protein of canine PF
Human PF is characterized by the presence of autoantibodies against Dsg1. Canine PF is believed to be counterpart of human PF because of the similarity in clinical manifestations and subcorneal lesions and the recognition of the 160 kDa proteins in canine PF sera by immunoblottings. More than 90% of sera from human PF patients were found to react with normal intercellular antigens by indirect immunofluorescence (IIF) and to show the good affinity for the recombinant Dsg1 proteins by ELISA. By contrast, most of canine PF sera did not react with normal substrates by IIF and did not bind to recombinant canine Dsg1 proteins. Part of the extracellular region was cloned and sequenced by Muller and Suter et al. and Nishifuji et al. prepared the recombinant proteins of the extracellular domain of canine Dsg1 and attempted to react the proteins with several sera of canine PF patients by the immunoblotting-immunoprecipitation method (IP-IB) however none of the sera recognized recombinant canine Dsg1.
Diseases causing non-autoimmune PF-like conditions
The infection of Trichophyton mentagrophytes has been reported to cause the acantholysis and subcorneal pustules in the hair follicle cells and epidermis like canine PF. As some cases of drug-induced reactions or erythema multiforme show acantholysis and subcorneal pustules histopathologically, these may be considered to be PF if there is sufficient evidences to suspect drug. In such cases, the direct immunofluorescence test (DIF) is often positive whereas IIF is rarely positive.
Recently, Amagai et al. reported that the target protein of the exotoxin secreted by Staphylococcus aureus is Dsg1 in human. In children suffering from the staphylococcal skin-scalded syndrome (SSSS), whereas the staphylococcal infection is known to cause pustular lesions on the skin, the target proteins of the exotoxin have remained unknown for a long time. In those children, the exotoxin digested Dsg1 eventually attracting neutrophils and inducing the formation of pustules. The same mechanism may occur in swine staphylococcal infection. Nishifuji et al. sequenced the exotoxin of Staphylococcus hyicus, produced a recombinant exotoxin and confirmed that the exotoxin may digest swine Dsg1 in vitro and in vivo. In dog, although the exotoxin produced by Staphylococcus intermedius has not known so far, it is not surprising event if some of the PF-like conditions may be caused by a mechanism to that described above.
Are there any other possible desmosomal proteins as the target protein of canine PF? Desmocollins 1 and 2?
Some of human IgA pemphigus may resemble to canine PF in terms of histopathological findings, as the finding of subcorneal pustules is the common to both. However, there are few reports stating that canine PF patient have IgA class autoantibodies and most of the autoantibodies are IgG, particularly IgG1 and IgG4. Desmocollins are thought to be the target protein of human IgA pemphigus. Recently, Aoki et al. cloned and sequenced desmocollins 1 and 2 and attempted to bind them to sera of canine PF by IIF and immunoblotting-immunoprecipitation (IB-IP) methods using IgG and IgA as the secondary antibodies, but the results were disappointing.
Is Dsg1 a target protein in both canine PF as well as human PF?
The lines of evidences presented above indicate the following possibilities:
1. Canine PF does not have autoantibodies, therefore, it is not an autoimmune diseases.
2. Canine PF may have autoantibodies against unknown intercellular proteins other than Dsg1 and Dsg3 and desmocollins, such as canine Dsg4.
3. There may be some variants of canine PF, some being true autoimmune diseases and some being otherwise, in spite of their having very similar clinical signs and histopathological findings?
4. There may be some factors interfering with the immunological reaction between antigens and autoantibodies in vitro.
Which domain of desmogleins is the critical for the occurrence of PF and PV?
In human PF and PV, the domain of Dsg that reacts with patient sera was studied in detail using chimeric proteins between Dsg 1 and 3.
Sekiguchi et al. generated Dsg1- and Dsg3-domain-swapped molecules and point-mutated Dsg3 molecules with Dsg1-specific residues by baculovirus expression. Using the predicted three-dimensional structure of classic cadherins as a model, they suggested that the dominant autoimmune epitopes in both PF and PV are found in the N-terminal adhesive surfaces of Dsgs.
Desmoglein 3 (Dsg3)
Dsg 3 is the target protein of human PV and it is believed that human and canine PV autoantibodies recognize the same target proteins based on the findings of Suter et al. and Iwasaki et al. that the 130 kDa protein in keratinocyte extract was recognized by a serum from canine PV patient. Later, Olivry et al. reported the same findings by large number of patient sera from canine PV. Recently, Nishifuji et al. cloned and sequenced canine Dsg3 and successfully expressed it in the baculovirus system. This recombinant Dsg3 absorbed canine and human autoantibodies, suggesting the target protein of canine PV is Dsg3.
Distribution of Dsg1, Dsg3 and pemphigus subtypes
Although it remains to be confirmed if the target protein of PF is indeed Dsg1, it is still the first candidate for the target protein of canine PF. In general, clinical signs of canine PF appear on the skin and rarely on the mucous membrane, whereas those of canine PV are seen mainly on the mucous membrane. The reason for the difference in clinical expression between those two diseases is not known. Aoki et al. studied the distribution and expression of both Dsgs 1 and 3 on the skin and mucous membrane using sera of human PF and PV patients, and found that Dsg1 was distributed mainly on the upper epidermis and was less concentrated in the upper and middle parts of the mucous membrane. In contrast, Dsg3 was distributed mainly throughout the mucous membrane. These results imply that in the presence of an autoantibody against Dsg1, the epidermis may be affected adversely, whereas the mucous membrane will remain intact due to the distribution of compensation of Dsg3. On the other hand, in the presence of an autoantibody against Dsg3, the epidermis will not affected by the paucity of antigens, and only the lower portion of the mucous membrane will be affected because of the paucity of compensation by Dsg1. The distribution of Dsgs1 and 3 is well correlated closely with the expression of the clinical signs of PF and PV.
1. Amagai M: Usefulness of enzyme-linked immunosorbent assay using recombinant desmogleins 1 and 3 for serodiagnosis of pemphigus. Br J Dermatol, 140: 351-357, 1999
2. Amagai M, Matsuyoshi N, Wang ZH, Andl C, Stanley JR: Toxin in bullous impetigo and staphylococcal scalded-skin syndrome targets desmoglein 1. Nat Med, Nov 6: 1275-1277, 2000
3. Aoki M, Amagai M, Nishifuji T, Nishikawa T, Iwasaki T: Distribution and expression of desmoglein 1 and 3 in canine skin and mucous membrane. Adv Vet Dermatol 4: 30-36, 2003
4. Iwasaki T, Shimizu M, Obata M: Detection of canine pemphigus foliaceus autoantigen by immunoblotting. Vet Immunol Immunopathol, 59: 1-10, 1997
5. Iwasaki T, Olivry T: Spontaneous canine model of pemphigus foliaceus. In: Animal Model of Human Inflammatory Skin Diseases. Edited by Chan LS, CRC press, New York, 309-320, 2003
6. Muller E, Caldelari R, Levine R, Kaplan S, Baron A, Rohrbach B, Wyder M, Balmer V, Suter MM: Cloning of canine Dsg1 and evidence for alternative polyhydration. J Invest Dermatol, 114:1211-1213, 2000
7. Nishifuji K, Amagai M, Nishikawa T, Iwasaki T: Production of recombinant extracellular domains of canine desmoglein 1 (Dsg1) by baculovirus expression. Vet Immunol Immunopathol, 95: 177-182, 2003
8. Nishifuji K, Ota K, Amagai M, Iwasaki T: Cloning of canine desmoglein 3 and immunoreactivity of serum antibodies in human and canine pemphigus vulgaris with its extracellular domains. J Dermatol Sci, 32: 181-191, 2003
9. Olivry T: Desmoglein-3 is a target autoantigen in spontaneous canine pemphigus vulgaris. Exp Dermatol, 12: 198-203, 2003
10. Olivry T: Spontaneous canine model of pemphigus vulgaris. In: Animal model of inflammatory skin diseases. Edited by Chan LS, CRC press, New York263-273, 2003
11. Sekiguchi M, @Futei Y, Fujii Y, Iwasaki T, Nishikawa T, Amagai M: J Immunol, 2002
12. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins. J Immunol, 167: 5439-48, @2001
13. White SD, Carlloti DN, Pin T, Bohnenberger T, Ihrke PJ, Monet E, Nishifuji K, Iwasaki T: Putative drug-related pemphigus foliaceus in four dogs. Vet Dermatol, 13: 195-202, 2002