Vet Dermatol 2014; 25: 471–e75

DOI: 10.1111/vde.12143

Serum autoantibody profiles of IgA, IgE and IgM in canine pemphigus foliaceus Petra Bizikova*†, Thierry Olivry*†, Lisa B. Mamo* and Stanley M. Dunston* *Department of Clinical Sciences and †Center for Comparative Medicine and Translational Research, College of Veterinary Medicine, North Carolina State University, 1060 William Moore Drive, Raleigh, NC 27607, USA Correspondence: Petra Bizikova, Department of Clinical Science, North Carolina State University, College of Veterinary Medicine, 1060 William Moore Drive, Raleigh, NC 27607, USA. E-mail: [email protected]

Background – Pemphigus foliaceus (PF) is the most common IgG-mediated autoimmune skin disease in dogs. Studies of human PF have revealed the presence of other antigen-specific autoantibody isotypes, thereby uncovering new avenues of investigation of the disease pathomechanism. Hypothesis/Objectives – The aim was to obtain information about the autoantibody isotype response in canine PF. Methods – Sera from 34 dogs with PF were tested for the presence of antikeratinocyte, anti-desmocollin-1 and anti-desmoglein-1 IgA, IgE and IgM using indirect immunofluorescence. Results – Using our indirect immunofluorescence technique, IgA, IgE and IgM autoreactivities were detected in six, one and zero of 34 sera from PF-affected dogs, respectively. Two of the six IgA-positive sera contained antikeratinocyte and anti-desmocollin-1 IgA, while the four remaining sera tested positive either for antikeratinocyte IgA (two of six) or for anti-desmocollin-1 IgA (two of six). A single serum contained anti-desmocollin-1 IgE. None of the six sera from healthy dogs contained detectable IgA, IgE or IgM autoantibodies. Conclusions and clinical importance – Our findings suggest that sera from dogs with PF rarely contain IgA or IgE autoantibodies at levels detectable by indirect immunofluorescence, while IgM autoreactivity appears not to be a feature of this disease. Considering these findings, it appears that canine PF is aetiologically and immunologically similar to that of the classic human PF, in which the IgG autoantibody response is also the predominant type.

Introduction Pemphigus foliaceus (PF) is a debilitating acantholytic pustular disease caused by IgG autoantibodies, which disrupt adhesion between keratinocytes.1 In most dogs with PF, these antibodies target desmocollin-1 (DSC1), one of several desmosomal cadherins involved in cell–cell adhesion of keratinocytes.2 Only a minority of dogs with PF will also have antibodies against another desmosomal cadherin, desmoglein-1 (DSG1), the major target autoantigen in human PF.3,4 With the exception of our knowledge about the dominant IgG subclass response in dogs with PF, which, as in humans, involves predominantly IgG4,5 little is known about any other antibody isotype response in canine PF. It has been established previously that the production of IgG4 is dependent on a T-helper-2 (Th2)-type cytokine milieu.6 The same cytokine milieu is also involved in IgE production,7 thus raising the possibility that dogs with PF might also produce antikeratinocyte-, DSC1- and/or DSG1-targeting IgE antibodies. Indeed, IgE antibodies targeting the major human PF autoantigen, DSG1, have been detected in ~80 and ~40% of sera of human individuals with endemic and nonendemic PF, respectively.8

Accepted 4 April 2014 Sources of Funding: This study was self-funded. Conflict of Interest: No conflicts of interest have been declared. © 2014 ESVD and ACVD, Veterinary Dermatology, 25, 471–e75.

Although IgA antibodies are rarely detected in humans with classic PF,9 IgA is the dominant antibody isotype response detected in a rare form of pemphigus called IgA pemphigus. Interestingly, subcorneal pustular dermatosis, a subset of IgA pemphigus with anti-DSC1 IgA autoantibodies, is characterized by a strong neutrophilic inflammatory response, which accompanies blister formation, a feature similar to that seen in canine PF.1,10 In addition, anti-DSC1 IgA antibodies were uncovered in atypical forms of pemphigus foliaceus where neutrophilic inflammation is also a prominent feature of blister formation.11–13 Finally, IgM is considered to be the result of primary antigen stimulation, and it is the first antibody produced before re-exposure causes antibody isotype switching.14 While seen only rarely in people with classic PF, anti-DSG1 IgM antibodies are detected in people with the endemic form of PF called fogo selvagem, as well as in healthy people living in the endemic region.15 This phenomenon is attributed to exposure to an environmental antigen, a sand fly salivary protein, which causes antigen mimicry and triggers the production of cross-reacting anti-DSG1 antibodies and fogo selvagem in genetically susceptible individuals.16 The role of molecular mimicry and the presence of circulating antigen-specific IgM in dogs with PF have not been investigated, although a possible association between flea allergy and PF was hypothesized in one study.17 This observation must be interpreted with caution, because there was a high prevalence of flea allergy in that area and at the time when the study was performed. 471

Bizikova et al.

In an effort to obtain information about the antibody isotype response in canine PF, sera from 34 dogs with PF were tested for the presence of antikeratinocyte, antiDSC1 and anti-DSG1 IgM, IgE and IgA antibodies using indirect immunofluorescence (IF).

Results Using the indirect IF technique on healthy canine footpad, buccal mucosa, DSC1- and/or DSG1-transfected cells, IgA autoreactivity was detected in six of 34 sera (18%) from PF-affected dogs (Table 1). Two of these sera tested positive for antikeratinocyte and anti-DSC1 IgA, while the four remaining sera tested positive either for antikeratinocyte IgA (two) or anti-DSC1 IgA (two). None of the sera contained anti-DSG1 IgA. All IgA-positive sera also contained antikeratinocyte and/or anti-DSC1 IgG antibodies (Table 1), suggesting that a true IgA pemphigus, as described in humans,10 was not detected among the selected dogs with PF in this study. None of the six healthy canine sera tested positive for antikeratinocyte, anti-DSC1 and anti-DSG1 IgA antibodies (Table 1). We were able to detect anti-DSC1 IgE autoantibodies in serum from only one dog with PF, which also had antikeratinocyte and anti-DSC1 IgG autoantibodies (Table 1). Immunoglubulin M autoreactivity was not confirmed in any of the tested samples. None of the six healthy canine sera tested positive for antikeratinocyte, anti-DSC1 or anti-DSG1 IgE and IgM antibodies (Table 1).

Materials and methods Canine sera Serum was collected from 34 dogs with PF diagnosed based on the presence of typical clinical signs and lesion distribution supported by typical microscopic (histological and/or cytological) signs compatible with this disease.1 All sera had been previously tested for antikeratinocyte, anti-DSC1 and anti-DSG1 IgG antibodies using indirect IF as described previously.2 The sera were then divided based on the presence or absence of antikeratinocyte, anti-DSC1 and anti-DSG1 IgG into four groups, as shown in Table 1. In addition, six sera from healthy dogs without detectable antikeratinocyte, anti-DSC1 and anti-DSG1 IgG were also included in the study (negative control sera).

Detection of circulating antikeratinocyte, anti-DSC1 and anti-DSG1 IgA, IgE and IgM by indirect immunofluorescence Detection of circulating antikeratinocyte IgA, IgE and IgM was performed by indirect IF on canine footpad and buccal mucosa tissues using a previously published protocol.18 The protocol was adapted to detect each specific antibody isotype by replacing the anti-canine IgG antibody with fluorescein isothiocyanate (FITC)-conjugated goat anticanine IgA at 1:200 dilution (Bethyl Labs, Montgomery, TX, USA), FITC-conjugated goat anti-canine IgE at 1:200 dilution (AbD Serotec, Hercules, CA, USA) or FITC-conjugated goat anti-canine IgM at 1:800 dilution (Bethyl Labs). All sera were tested twice for each antibody isotype at 1:10, 1:20, 1:50 and 1:100 dilutions. Detection of anti-DSC1 and anti-DSG1 IgA, IgE and IgM was performed by indirect IF on DSC1- and DSG1-tranfected 293T cells. The 293T cells were transfected using a previously published protocol.2 Nontransfected 293T cells served as negative controls for each tested serum and antibody isotype. Indirect IF followed a previously described protocol,2 but was adapted to detect each specific isotype by replacing the anti-canine IgG antibody with relevant antibodies as described above. All sera were tested twice for each antibody isotype at 1:10, 1:20, 1:50 and 1:100 dilutions. Canine spleen and mucosal tissues were included as plasma cell-containing positive controls for the secondary, isotype-specific antibodies.

Discussion Using an indirect IF technique established in our laboratory, we were able to detect IgA and IgE autoreactivity in six and one of 34 sera from PF-affected dogs, respectively, while IgM autoreactivity was not detected in any of these sera. None of the six healthy canine sera tested positive for IgA, IgE or IgM autoantibodies. The coexistence of IgG and IgA autoantibodies is not unique to canine PF, because similar data had been reported for its human counterpart.9 Two sera that tested positive for anti-DSC1 IgA antibodies on DSC1-transfected cells did not test positive for antikeratinocyte IgA on footpad and buccal mucosa tissues. This could be the result of a low IgA autoantibody titre, which could be detected only by a more sensitive method using DSC1-transfected cells that highly overexpress the target

Table 1. Summary of IgA, IgE and IgM immunoreactivity of canine pemphigus foliaceus sera and sera from healthy dogs divided based on their IgG autoreactivity Antikeratinocyte antibodies

Pemphigus foliaceus (34)

Healthy dogs (6)

Anti-desmocollin-1 antibodies

Anti-desmoglein-1 antibodies

IgG autoreactivity (dogs per group)

IgA

IgE

IgM

IgA

IgE

IgM

IgA

IgE

IgM

Antikeratinocyte IgG+, DSC1+/DSG1 (11) Antikeratinocyte IgG+, DSC1 /DSG1 (9) Antikeratinocyte IgG , DSC1+/DSG1 (6) Antikeratinocyte IgG , DSC1 /DSG1 (8)

3

0

0

2

1

0

0

0

0

1

0

0

1

0

0

0

0

0

0

0

0

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Antikeratinocyte IgG , DSC1 /DSG1 (6)

Abbreviations: DSC1, desmocollin-1; DSG1, desmoglein-1. Note that reactivity within individual sera, as listed in the table, is not mutually exclusive. Two sera were positive for both anti-keratinocyte and anti-DSC1 IgA, such that a total of six sera exhibited IgA reactivity to individual antigens.

472

© 2014 ESVD and ACVD, Veterinary Dermatology, 25, 471–e75.

Antibody isotypes of pemphigus foliaceus

antigen. In contrast, two other sera were positive for antikeratinocyte IgA autoantibodies using canine footpad and buccal mucosa but lacked positivity on DSC1- and DSG1tranfected cells, suggesting that these two sera might have had IgA autoantibodies binding to keratinocyte antigen(s) other than the DSC1 and DSG1. Although the presence of circulating serum IgA autoantibodies in canine PF had not been published before, a review of the literature uncovered a study evaluating the deposition of antikeratinocyte autoantibodies in the skin of dogs with pemphigus. Results of this study were similar to ours, and detected the concurrent presence of IgG and IgA autoantibodies in two of 26 and two of nine dogs with PF and pemphigus erythematosus, respectively.19 Given that there are no data about circulating, antigenspecific IgE or IgM autoantibodies in sera from dogs with pemphigus with which we could compare our results, we can only speculate about the reason for the low positivity rate observed in our study. It is possible that canine PF is predominantly an IgG-mediated disease and only rare cases contain sufficient serum levels of other, antigenspecific antibody isotypes, which can be detected by an indirect IF. It is possible that a more sensitive technique, for example enzyme-linked immunosorbent assay, could exhibit a higher positive detection rate. Unfortunately, a DSC1- or DSG1-specific enzyme-linked immunosorbent assay is currently not available for dogs. In humans, the triple response of anti-DSG1 IgG, IgM and IgE is markedly more common in the endemic Brazilian pemphigus fogo selvagem than in the classic PF diagnosed in patients from the USA and Japan. This observation has led to the hypothesis that different pathomechanisms are involved in these two entities.8 Indeed, a persistent antigenic stimulation by an environmental allergen, possibly a sand fly salivary antigen, is likely to be one of the triggers causing the endemic disease in genetically predisposed individuals.16 Considering our findings, it is possible that canine PF is aetiologically and immunologically more similar to that of the classic human PF rather than fogo selvagem, because most canine PF sera appear to lack multiple detectable autoantibody isotypes. Although previous information about circulating antigenspecific IgE and IgM in canine pemphigus is unavailable, tissue-bound antikeratinocyte IgM autoantibodies were detected in the skin of only two of 26 dogs with PF and two of nine dogs with pemphigus erythematosus.19 This finding is not unusual, because direct IF using affected skin from a dog with erosions, pustules and crusts may contain intercellular serum deposits due to the surrounding tissue inflammation. As a result, intercellular antibody deposits can be found in biopsy specimens obtained from dermatoses other than PF.20 In summary, our findings suggest that sera from dogs with PF rarely contain antikeratinocyte or anti-DSC1 IgA or IgE autoantibodies at levels detectable by indirect IF, while IgM autoreactivity appears not to be a feature of this disease. A true IgA pemphigus without concurrent IgG autoantibodies was not identified among the cases included. Considering the limitations of the indirect IF, future studies should re-evaluate the isotype antibody profile of canine PF using a more sensitive method of detection, such as an antigen-specific enzyme-linked © 2014 ESVD and ACVD, Veterinary Dermatology, 25, 471–e75.

immunosorbent assay, a technique currently unavailable for use with canine sera.

Acknowledgements The authors would like to thank Keith Linder for reviewing the manuscript and for his constructive comments.

References 1. Olivry T. A review of autoimmune skin diseases in domestic animals: I – Superficial pemphigus. Vet Dermatol 2006; 17: 291–305. 2. Bizikova P, Dean GA, Hashimoto T et al. Cloning and establishment of canine desmocollin-1 as a major autoantigen in canine pemphigus foliaceus. Vet Immunol Immunopathol 2012; 149: 197–207. 3. Olivry T, LaVoy A, Dunston SM et al. Desmoglein-1 is a minor autoantigen in dogs with pemphigus foliaceus. Vet Immunol Immunopathol 2006; 110: 245–255. 4. Joly P, Litrowski N. Pemphigus group (vulgaris, vegetans, foliaceus, herpetiformis, brasiliensis). Clin Dermatol 2011; 29: 432–436. 5. Olivry T, Dunston SM, Walker RH et al. Investigations on the nature and pathogenicity of circulating antikeratinocyte antibodies in dogs with pemphigus foliaceus. Vet Dermatol 2009; 20: 42–50. 6. Aalberse RC, Stapel SO, Schuurman J et al. Immunoglobulin G4: an odd antibody. Clin Exp Allergy 2009; 39: 469–477. 7. Geha RS, Jabara HH, Brodeur SR. The regulation of immunoglobulin E class-switch recombination. Nat Rev Immunol 2003; 3: 721–732. 8. Qian Y, Prisayanh P, Andraca E et al. IgE, IgM, and IgG4 antidesmoglein 1 autoantibody profile in endemic pemphigus foliaceus (fogo selvagem). J Invest Dermatol 2011; 131: 985–987. 9. Mentink LF, de Jong MC, Kloosterhuis GJ et al. Coexistence of IgA antibodies to desmogleins 1 and 3 in pemphigus vulgaris, pemphigus foliaceus and paraneoplastic pemphigus. Br J Dermatol 2007; 156: 635–641. 10. Tsuruta D, Ishii N, Hamada T et al. IgA pemphigus. Clin Dermatol 2011; 29: 437–442. 11. Hisamatsu Y, Amagai M, Garrod DR et al. The detection of IgG and IgA autoantibodies to desmocollins 1–3 by enzyme-linked immunosorbent assays using baculovirus-expressed proteins, in atypical pemphigus but not in typical pemphigus. Br J Dermatol 2004; 151: 73–83. 12. Heng A, Nwaneshiudu A, Hashimoto T et al. Intraepidermal neutrophilic IgA/IgG antidesmocollin 1 pemphigus. Br J Dermatol 2006; 154: 1018–1020. €ller R, Heber B, Hashimoto T et al. Autoantibodies against 13. Mu desmocollins in European patients with pemphigus. Clin Exp Dermatol 2009; 34: 898–903. 14. Murphy K, Travers P, Walport M et al. Janeway’s Immunobiology. New York: Garland Science, 2012. 15. Diaz LA, Prisayanh PS, Dasher DA et al. The IgM anti-desmoglein 1 response distinguishes Brazilian pemphigus foliaceus (fogo selvagem) from other forms of pemphigus. J Invest Dermatol 2008; 128: 667–675. 16. Qian Y, Jeong JS, Maldonado M et al. Cutting edge: Brazilian pemphigus foliaceus anti-desmoglein 1 autoantibodies cross-react with sand fly salivary LJM11 antigen. J Immunol 2012; 189: 1535–1539. 17. Pascal A, Shiebert J, Ihrke P. Seasonality and environmental risk factors for pemphigus foliaceus in animals: a retrospective study of 83 cases presented to the veterinary medical teaching hospital, University of California Davis from 1976 and 1994 (abstract). In: Proceedings of the American Academy of Veterinary Dermatology & American College of Veterinary Dermatology. Santa Fe, NM, USA, 1995; 24–25. 18. Bizikova P, Linder KE, Olivry T. Immunomapping of desmosomal and nondesmosomal adhesion molecules in healthy canine foot-

473

Bizikova et al.

pad, haired skin and buccal mucosal epithelia: comparison with canine pemphigus foliaceus serum immunoglobulin G staining patterns. Vet Dermatol 2011; 22: 132–142. 19. Scott DW, Walton DK, Slater MR et al. Immune-mediated dermatoses in domestic animals: Ten years after - part I. Compend Contin Educ Small Anim Pract 1987; 9: 424–435.

20. Werner LL, Brown KA, Halliwell REW. Diagnosis of autoimmune skin disease in the dog: correlation between histopathologic, direct immunofluorescent and clinical findings. Vet Immunol Immunopathol 1983; 5: 47–64.

sume  Re  (PF) est la dermatite auto-immune me die e par les IgG la plus fre quente Contexte – Le pemphigus foliace tudes sur le PF humain ont re ve  le  la pre sence d’autres isotypes d’auto-anticorps spe cifchez le chien. Les e nes. iques d’antige ses/Objectifs – Le but e tait d’obtenir des informations sur l’isotype de re ponse d’auto-anticorps Hypothe dans le PF canin. thodes – Les sera de 34 chiens atteints de PF ont e  te  teste s pour la pre sence d’IgA, IgE et IgM antiMe ratinocytes, anti-desmocolline-1 et anti-desmogle ine par immunofluorescence indirecte. ke sultats – A l’aide de notre technique d’immunofluorescence indirecte, les auto-re activite s IgA, IgE et Re  te  de tecte es respectivement pour six, un et ze ro des 34 sera de chiens atteints de PF. Deux des IgM ont e ratinocytes et anti-desmocolline-1 alors que les quatre ausix sera IgA-positifs contenaient des IgA anti-ke s e taient positifs soit pour les IgA anti-ke ratinocytes (deux sur six) et pour les IgA anti-destres sera teste mocolline (deux sur six). Un seul serum contenait des IgE anti-desmocolline. Aucun des six sera issus de celables. chiens sains de contenaient d’auto-anticorps IgA, IgG et IgM de es sugge rent que les sera des chiens atteints de PF Conclusions et importance clinique – Nos donne celables par immunofluorescence indicontiennent rarement des auto-anticorps IgA et IgE  a des niveaux de activite  IgM semble ne pas e ^tre une caracte ristique de cette maladie. Ainsi, il semrecte alors que l’auto-re tiologiquement et immunologiquement semblable au classique PF humain dans ble que le PF canin est e galement le type pre dominent. lequel les auto-anticorps IgG sont e Resumen  n – el pe nfigo foliaceo (PF) es la enfermedad autoinmune m n mediada por IgG en perIntroduccio as comu nfigo foliaceo en humanos han desvelado la presencia de otros y isotipos de autoanros. Los estudios de pe n del ticuerpos especıficos del antıgeno, por lo tanto abriendo nuevas avenidas en la investigacio patomecanismo de la enfermedad.  tesis/Objetivos – el propo sito fue obtener informacio n acerca de isotipos de inmunoglobulina en la Hipo nfigo foli respuesta de autoanticuerpos en el pe aceo canino. todos – el suero de 34 perros con pe nfigo foli Me aceo fue probado para la presencia de anticuerpos antiqueratinocitos, antidesmocolina-1 y antidesmogleina-1, de los isotipos IgA, IgE e IgM utilizando inmunofluorescencia indirecta. ma de inmunofluorescencia indirecta, se detecto  autoreactividad de Resultados – utilizando nuestro siste nfigo foli IgA, y IgE e IgM en 6, uno y ninguno de los 34 sueros de perros afectados con pe aceo, respectivamente. Dos de los seis sueros positivos a IgA contenıan IgA antiqueratinocitos y antidesmocolina-1, mientras que los 4 sueros restantes fueron positivos para IgA antiqueratinocitos (dos de seis) o para IgA lo suero contenıa IgE antidesmocolina-1. Ninguno los seis sueros de antidesmocolina-1 (dos de seis). Un so los perros sanos contenıa niveles detectables de autoanticuerpos IgA, IgE o IgM. nfigo Conclusiones e importancia clınica – nuestros hallazgos sugieren que el suero de perros con pe foliaceo raramente contiene autoanticuerpos IgA o IgE a niveles detectables mediante inmunofluorescencia indirecta, mientras que la autoreactividad de IgM parece no ser una caracterıstica de esta enfermedad. nfigo foli gicamente e inmunolo gicaConsiderando estos hallazgos, parece que el pe aceo canino es etiolo nfigo foli mente similar a la forma clasica de pe aceo en humanos, en la cual la respuesta de autoanticuerpos n el tipo predominante. IgG es tambie Zusammenfassung Hintergrund – Pemphigus foliaceus (PF) ist die h€ aufigste durch IgG-mediierte Autoimmunerkrankung bei €ber PF des Menschen haben das Vorkommen von anderen Antigen-spezifischen AuHunden. Studien u €rper Isotypen festgestellt, wobei neue Wege gezeigt wurden mit denen Pathomechanismen von toantiko €nnen. Krankheiten untersucht werden ko €rper Isoty€ber die Antwort der Autoantiko Hypothese/Ziele – Das Ziel dieser Studie war es, Information u pen beim PF des Hundes zu erlangen. Methoden – Sera von 34 Hunden mit PF wurden auf das Vorkommen von Antikeratinozyten, Anti-desmocollin-1 und Anti-Desmoglein-1 IgA, IgE und IgM mittels indirekter Immunfluoreszenz getestet. Ergebnisse – Mithilfe unserer indirekten Immunfluoreszenztechnik wurden IgA, IgE bzw. IgM Autoreaktionen in sechs, einem bzw. keinem von 34 Sera von PF-betroffenen Hunden gefunden. Zwei der sechs IgApositiven Sera wiesen Antikeratinzyten und Anti-Desmocollin-1 IgA auf, w€ ahrend die vier verbleibenden 474

© 2014 ESVD and ACVD, Veterinary Dermatology, 25, 471–e75.

Antibody isotypes of pemphigus foliaceus

€r Antikeratinozyten IgA (zwei von sechs) oder fu €r Anti-Desmocollin-1 IgA (zwei von sechs) Sera entweder fu positiv waren. Ein einziges Serum wies Anti-Desmocollin-1 IgE auf. Keines der sechs Sera von gesunden €rper auf. Hunden wies messbare IgA, IgE oder IgM Autoantiko Schlussfolgerungen und klinische Bedeutung – Unsere Ergebnisse zeigen, dass in Sera von Hunden €rper mittels indirekter Immunfluoreszenz gemessen werden ko €nnen, mit PF selten IgA oder IgE Autoantiko €rfte. In Anbetracht dieser Ergebnisse w€ahrend IgM Autoreaktivit€at kein Merkmal dieser Erkrankung sein du scheint der PF des Hundes €atiologisch und immunologisch € ahnlich zu sein, wie der klassische PF des €rperantwort ebenfalls die vorherrschende Antwort darstellt. Menschen, in welchem die IgG Autoantiko

© 2014 ESVD and ACVD, Veterinary Dermatology, 25, 471–e75.

e75