Portal de Periódicos da CAPES

Epidermal Transglutaminase Deposits in Perilesional and Uninvolved Skin in Patients with Dermatitis Herpetiformis

2007; Elsevier BV; Volume: 127; Issue: 5 Linguagem: Inglês

10.1038/sj.jid.5700682

1523-1747

Matthew R. Donaldson, John J. Zone, Linda Schmidt, Ted B. Taylor, Susan L. Neuhausen, Christopher Hull, Laurence J. Meyer,

Skin and Cellular Biology Research

direct immunofluorescence dermatitis herpetiformis endomysial antibody gluten-free diet transglutaminase 2 or tissue transglutaminase transglutaminase 3 or epidermal transglutaminase TO THE EDITOR Dermatitis herpetiformis (DH) is a disease precipitated by ingestion of gluten and characterized by IgA deposits in the dermal papillae (Nicolas et al., 2003Nicolas M.E. Krause P.K. Gibson L.E. Murray J.A. Dermatitis herpetiformis.Int J Dermatol. 2003; 42: 588-600Crossref PubMed Scopus (96) Google Scholar). Like celiac disease, DH is one presentation of gluten-sensitive enteropathy (Zone, 2005Zone J.J. Skin manifestations of celiac disease.Gastroenterology. 2005; 128: S87-91Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). In susceptible individuals, ingestion of gluten leads to elaboration of antibodies against gliadin and transglutaminase 2 (TG2) (Oxentenko and Murray, 2003Oxentenko A.S. Murray J.A. Celiac disease and dermatitis herpetiformis: the spectrum of gluten-sensitive enteropathy.Int J Dermatol. 2003; 42: 585-587Crossref PubMed Scopus (45) Google Scholar; Alaedini and Green, 2005Alaedini A. Green P.H. Narrative review: celiac disease: understanding a complex autoimmune disorder.Ann Intern Med. 2005; 142: 289-298Crossref PubMed Scopus (262) Google Scholar; Zone, 2005Zone J.J. Skin manifestations of celiac disease.Gastroenterology. 2005; 128: S87-91Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). TG2, one of nine homologous transglutaminases, is an autoantigen in celiac disease (Dieterich et al., 1997Dieterich W. Ehnis T. Bauer M. Donner P. Volta U. Riecken E.O. et al.Identification of tissue transglutaminase as the autoantigen of celiac disease.Nat Med. 1997; 3: 797-801Crossref PubMed Scopus (1791) Google Scholar; Lorand and Graham, 2003Lorand L. Graham R.M. Transglutaminases: crosslinking enzymes with pleiotropic functions.Nat Rev Mol Cell Biol. 2003; 4: 140-156Crossref PubMed Scopus (1211) Google Scholar). Serum antibodies against TG2 can be sensitive and specific for DH as well as celiac disease (Dieterich et al., 1999Dieterich W. Laag E. Bruckner-Tuderman L. Reunala T. Karpati S. Zagoni T. et al.Antibodies to tissue transglutaminase as serologic markers in patients with dermatitis herpetiformis.J Invest Dermatol. 1999; 113: 133-136Crossref PubMed Scopus (189) Google Scholar). Sardy et al., 2002Sardy M. Karpati S. Merkl B. Paulsson M. Smyth N. Epidermal transglutaminase (TGase 3) is the autoantigen of dermatitis herpetiformis.J Exp Med. 2002; 195: 747-757Crossref PubMed Scopus (439) Google Scholar hypothesized that an autoantigen related to TG2 could explain the presentation of DH and reported that epidermal transglutaminase (TG3) deposits with IgA in perilesional biopsies in patients with DH. IgA deposits can occur in regions that are not clinically involved (Fry et al., 1978Fry L. Haffenden G. Wojnarowska F. Thompson B.R. Seah P.P. IgA and C3 complement in the uninvolved skin in dermatitis herpetiformis after gluten withdrawal.Br J Dermatol. 1978; 99: 31-37Crossref PubMed Scopus (35) Google Scholar; Zone et al., 1996Zone J.J. Meyer L.J. Petersen M.J. Deposition of granular IgA relative to clinical lesions in dermatitis herpetiformis.Arch Dermatol. 1996; 132: 912-918Crossref PubMed Google Scholar). To confirm the involvement of TG3 in DH and to determine whether TG3 deposits are restricted to perilesional skin, we have raised a new goat antibody against human TG3. In a series of nine DH patients with active disease, we compared direct immunofluorescence (DIF) for IgA and TG3 in perilesional and uninvolved skin. We also compared serology for TG2 and TG3 and results from DIF for IgA and TG3 in the context of a gluten-free diet (GFD). The protocol was approved by our Institutional Review Board, patients consented in writing, and we adhered to the Declaration of Helsinki Principles. All patients had a diagnosis of DH established by skin biopsy showing granular IgA and had active disease. Adherence to GFD was reported by patients as strict or poor. Serum was analyzed for IgA antibodies against TG2, TG3, and endomysium endomysial antibody (EMA). Semiquantitative detection of anti-TG2 and -TG3 were performed using ELISA (INOVA Diagnostics, San Diego, CA, and Immunodiagnostik AG, Bensheim, Germany). EMA were detected using indirect immunofluorescence on monkey esophagus (University of California at Davis) (Unsworth, 1996Unsworth D.J. ACP Broadsheet No 149: September 1996. Serological diagnosis of gluten sensitive enteropathy.J Clin Pathol. 1996; 49: 704-711Crossref PubMed Scopus (46) Google Scholar). Patient age, sex, family history, adherence to GFD, dapsone use, and serologies for antibodies against TG2, TG3, and endomysium are represented in Table 1. Five patients reported strict adherence to a GFD. Among patients adhering to a strict GFD, four of five (80%) were negative by IgA ELISA for anti-TG2 antibodies and EMA. All four patients reporting poor GFD adherence had anti-TG2 antibodies and two of four were positive by EMA. All five patients in the strict GFD group were seronegative for anti-TG3 antibodies and all four patients in the poor GFD group were positive. The two patients with negative EMA had the lowest values for anti-TG2.Table 1Patient characteristics and serologyPatientSexAgeFHxGFD1All patients taking dapsone stopped medication for 48–72hours before biopsy and evaluation.DapsoneAnti TG2-IgA unitsIgA-EMA, titerIgA anti-TG3DH1M61NoStrictNo606DH2F28NoStrictNo605DH3F54YesStrict100 mg705DH4M29NoStrict200 mg1103DH5M72YesStrict25 mg671:8014DH6M82YesPoorNo25037DH7M61NoPoor25 mg36020DH8F54YesPoor200 mg1221:128020DH9M66NoPoorNo>1921:128042EMA, endomysial antibody; F, female; GFD, gluten-free diet; M, male.1 All patients taking dapsone stopped medication for 48–72 hours before biopsy and evaluation. Open table in a new tab EMA, endomysial antibody; F, female; GFD, gluten-free diet; M, male. Purified TG3 expressed from a baculovirus system was used to produce polyclonal antibodies in goats (Ahvazi and Steinert, 2003Ahvazi B. Steinert P.M. A model for the reaction mechanism of the transglutaminase 3 enzyme.Exp Mol Med. 2003; 35: 228-242Crossref PubMed Scopus (30) Google Scholar). Purified TG3 was a kind gift from Dr Bijan Ahazi. To confirm reactivity and specificity, Western blots were performed using the baculovirus-produced TG3 as well as separately cloned TG2 and TG3. Complementary DNAs for TG3 and TG2 were isolated by RT-PCR from human keratinocytes using primers published by Sardy et al., and were sequence confirmed. Complementary DNA was cloned in the Invitrogen pSecTag2/Hygro B vector, lysates were purified using a Ni2+ column (Invitrogen, Carlsbad, CA, USA), electrophoresed, and transferred to nitrocellulose blots. IgG was prepared using a Protein G column. Western blots confirmed a high titer against both TG3 clones. Specificity was demonstrated by lack of reactivity against non-transfected human embryonic kidney lysate, recombinant TG2 (Figure 1), and serum and secretory (colostral) IgA (not shown). Further specificity was established by lack of reactivity on IIF when reacted with skin from three patients with linear IgA bullous dermatosis (Figure 2e and f), two with Henoch–Shönlein purpura and human tonsil (not shown).Figure 2DIF in for IgA and TG3. (a–d) Specimens are from patient DH5. (a) Uninvolved skin showing 2+ DIF for IgA. (b) Uninvolved skin showing 1+ DIF for TG3. (c) Perilesional skin showing 3+ DIF by IgA. (d) Perilesional skin showing 2+ granular DIF by TG3 and some staining of the stratum corneum. (e, f) Specimens from a single biopsy of an LABD patient, although not adjacent cuts. (e) DIF for IgA showing a linear pattern. (f) No reaction with anti-TG3. Bar=0.05 mm.View Large Image Figure ViewerDownload (PPT) Sites of biopsy were chosen as non-erythematous skin adjacent to lesions (perilesional) and uninvolved skin 5 cm away. Four sections of each biopsy specimen were incubated with sheep anti-IgA FITC (1:40), goat IgG anti-TG3 (1:50), or preimmune goat IgG (1:50) for 30–60 minutes and washed. Anti-TG3 and preimmune control slides were incubated with FITC swine anti-goat IgG antibody (1:100) for 30 minutes before washing again. A single-blinded observer (J.J.Z.) graded all specimens by a semiquantitative scale (0–3+) (Zone et al., 1996Zone J.J. Meyer L.J. Petersen M.J. Deposition of granular IgA relative to clinical lesions in dermatitis herpetiformis.Arch Dermatol. 1996; 132: 912-918Crossref PubMed Google Scholar). Table 2 shows semiquantitative immunofluorescence values assigned to specimens from each of the nine patients. Typical staining patterns are shown in Figure 1a–d. All patients had granular deposits of IgA in the dermal papillae. Eight of nine had overlapping deposits of TG3. Only one specimen revealed IgA and TG3 deposition in vessel walls in the papillary dermis. TG3 showed staining of variable intensity of the upper layers of the epidermis, shown in Figure 1d. No IgA or TG3were seen in any specimen within the stratum basale.Table 2Semiquantitative DIF for perilesional and uninvolved skin biopsiesPerilesionalUninvolvedPatientIgATg3NegIgATg3NegDH12+DP1+DP02+DP1+DP0DH21+DP00000DH32+DP, DD1+DP, DD02+DP, DD1+DD0DH41+DP1+DP01+BM, VW1+BM, VW0DH53+DP continuous2+DP continuous02+DP, DD1+DP, DD0DH63+DP continuous2+DP02+DP continuous1+DP0DH72+DP1+DP01+DP1+DP0DH82+DP1+DP01+DP1+DP0DH93+DP continuous2+DP continuous03+DP2+DP0BM, basement membrane; DD, deep dermis, DIF, direct immunofluorescence; DP, dermal papillae, VW, vessel walls. Open table in a new tab BM, basement membrane; DD, deep dermis, DIF, direct immunofluorescence; DP, dermal papillae, VW, vessel walls. IgA DIF was more intense than TG3 in 13 of 18 specimens and equivalent in the remaining five. IgA and TG3 DIF were greater with poor GFD adherence. The only patient negative for TG3 was on a strict GFD and had sparse IgA in perilesional skin only. All five specimens from control subjects were negative for IgA and TG3 immunofluorescence, and the three LABD and two HSP patients were positive for IgA but negative for TG3. Patients with DH have TG3 in the papillary dermis overlapping with the deposits of IgA. Using a new antibody raised in goat with specificity confirmed using a separately cloned recombinant protein, we have confirmed findings of Sardy et al. suggesting TG3 is the autoantigen of DH. We also found TG3 deposits in uninvolved skin at least 5 cm away from lesions suggesting that factors beyond these complexes are necessary for DH eruptions. IgA deposits were seen in all specimens where TG3 was found, suggesting TG3 is bound by autoantibodies as the mechanism of deposition. TG3 was not been found in the dermis in the absence of IgA. The intensity of IgA by DIF roughly correlated with the intensity of staining for TG3. As TG3 is strongly expressed in the upper epidermis (Sardy et al., 2002Sardy M. Karpati S. Merkl B. Paulsson M. Smyth N. Epidermal transglutaminase (TGase 3) is the autoantigen of dermatitis herpetiformis.J Exp Med. 2002; 195: 747-757Crossref PubMed Scopus (439) Google Scholar), it is reasonable to suspect that in regions of trauma, it may be released from the epidermal keratinocytes and diffuse through the basement membrane. Circulating autoantibodies could then bind this antigen and form characteristic deposits. An alternate explanation involves circulating immune complexes that deposit at the dermoepidermal junction (Sardy et al., 2002Sardy M. Karpati S. Merkl B. Paulsson M. Smyth N. Epidermal transglutaminase (TGase 3) is the autoantigen of dermatitis herpetiformis.J Exp Med. 2002; 195: 747-757Crossref PubMed Scopus (439) Google Scholar; Preisz et al., 2005Preisz K. Sardy M. Horvath A. Karpati S. Immunoglobulin, complement and epidermal transglutaminase deposition in the cutaneous vessels in dermatitis herpetiformis.J Eur Acad Dermatol Venereol. 2005; 19: 74-79Crossref PubMed Scopus (50) Google Scholar), although we saw only one such involved vessel. Prior reports demonstrated IgA nephropathy in DH patients (Helin et al., 1983Helin H. Mustonen J. Reunala T. Pasternack A. IgA nephropathy associated with celiac disease and dermatitis herpetiformis.Arch Pathol Lab Med. 1983; 107: 324-327PubMed Google Scholar; Reunala et al., 1983Reunala T. Helin H. Pasternack A. Linder E. Kalimo K. Renal involvement and circulating immune complexes in dermatitis herpetiformis.J Am Acad Dermatol. 1983; 9: 219-223Abstract Full Text PDF PubMed Scopus (25) Google Scholar). Although circulating complexes could cause this pattern, such deposits could also result from a reaction against kidney TG3 (Sardy et al., 2002Sardy M. Karpati S. Merkl B. Paulsson M. Smyth N. Epidermal transglutaminase (TGase 3) is the autoantigen of dermatitis herpetiformis.J Exp Med. 2002; 195: 747-757Crossref PubMed Scopus (439) Google Scholar). Reported sensitivity of anti-TG2 ELISA and EMA for DH has varied (Chorzelski et al., 1986Chorzelski T.P. Jablonska S. Chadzynska M. Maciejowska E. Sulej J. IgA endomysium antibody in children with dermatitis herpetiformis treated with gluten-free diet.Pediatr Dermatol. 1986; 3: 291-294Crossref PubMed Scopus (10) Google Scholar; Reunala et al., 1987Reunala T. Chorzelski T.P. Viander M. Sulej J. Vainio E. Kumar V. et al.IgA anti-endomysial antibodies in dermatitis herpetiformis: correlation with jejunal morphology, gluten-free diet and anti-gliadin antibodies.Br J Dermatol. 1987; 117: 185-191Crossref PubMed Scopus (44) Google Scholar; Volta et al., 1992Volta U. Molinaro N. De Franchis R. Forzenigo L. Landoni M. Fratangelo D. et al.Correlation between IgA antiendomysial antibodies and subtotal villous atrophy in dermatitis herpetiformis.J Clin Gastroenterol. 1992; 14: 298-301Crossref PubMed Scopus (31) Google Scholar; Koop et al., 2000Koop I. Ilchmann R. Izzi L. Adragna A. Koop H. Barthelmes H. Detection of autoantibodies against tissue transglutaminase in patients with celiac disease and dermatitis herpetiformis.Am J Gastroenterol. 2000; 95: 2009-2014Crossref PubMed Scopus (83) Google Scholar; Sardy et al., 2000Sardy M. Karpati S. Peterfy F. Rasky K. Tomsits E. Zagoni T. et al.Comparison of a tissue transglutaminase ELISA with the endomysium antibody test in the diagnosis of gluten-sensitive enteropathy.Z Gastroenterol. 2000; 38: 357-364Crossref PubMed Scopus (23) Google Scholar). Studies correlate positive serology with more severe intestinal damage (Volta et al., 1992Volta U. Molinaro N. De Franchis R. Forzenigo L. Landoni M. Fratangelo D. et al.Correlation between IgA antiendomysial antibodies and subtotal villous atrophy in dermatitis herpetiformis.J Clin Gastroenterol. 1992; 14: 298-301Crossref PubMed Scopus (31) Google Scholar) and seronegativity with adherence to GFD (Peters and McEvoy, 1989Peters M.S. McEvoy M.T. IgA antiendomysial antibodies in dermatitis herpetiformis.J Am Acad Dermatol. 1989; 21: 1225-1231Abstract Full Text PDF PubMed Scopus (30) Google Scholar; Volta et al., 1992Volta U. Molinaro N. De Franchis R. Forzenigo L. Landoni M. Fratangelo D. et al.Correlation between IgA antiendomysial antibodies and subtotal villous atrophy in dermatitis herpetiformis.J Clin Gastroenterol. 1992; 14: 298-301Crossref PubMed Scopus (31) Google Scholar). Serology by IgA ELISA for anti-TG3 and TG2 antibodies correlated with adherence to GFD. Interestingly, EMA did not correlate tightly with TG3 or TG2 (Table 1). Mean DIF was greater for IgA and TG3 in patients seropositive for anti-TG2 and TG3 antibodies and reporting poor adherence to GFD, but a definitive relationship between serology and immune complex deposition cannot be established from this small patient series. The authors state no conflict of interest. We thank patients for participating in this study, Dr Kristin Leiferman and Marjorie Campbell from the immunodermatology laboratory at the University of Utah for technical assistance, and Wyatt Horsley for assistance with biopsies. This work was supported by NIH Grant DK50678 to SN, a Dermatology Foundation grant to CMH.