Portal de Periódicos da CAPES

Pathogenicity and Epitope Characteristics of Anti-Desmoglein-1 from Pemphigus Foliaceus Patients Expressing Only IgG1 Autoantibodies

2003; Elsevier BV; Volume: 121; Issue: 6 Linguagem: Inglês

10.1111/j.1523-1747.2003.12608.x

1523-1747

Mary K. Hacker-Foegen, Marleen M. Janson, Masayuki Amagai, Janet A. Fairley, Mong-Shang Lin,

Coagulation, Bradykinin, Polyphosphates, and Angioedema

Pemphigus foliaceus (PF) is an antibody-mediated autoimmune disorder with IgG1 and IgG4 as the predominant subclasses of autoantibodies against a desmosomal glycoprotein, desmoglein-1 (Dsg1). Previously, we found that the IgG4 anti-Dsg1 autoantibodies only recognize a conformational epitope(s), whereas the IgG1 autoantibodies recognize both conformational and linear epitopes but do not display pathogenicity in the passive transfer animal model. The purpose of this study was to analyze the epitopes recognized by autoanti-bodies from a subset of PF patients who only express anti-Dsg1 of the IgG1 isotype throughout the course of their diseases and to further characterize the pathogenicity of their IgG1 anti-Dsg1. We found that IgG1 auto-antibodies in this subset of PF patients, similar to IgG4 autoantibodies from other PF patients, are able to bind both human and mouse skin and induce the experimental PF in mice. Moreover, a detailed epitope mapping reveals that the conformational epitopes recognized by IgG1 autoantibodies from these PF patients are restricted to the first 161 amino acids of Dsg1, whereas the linear epitopes are spread throughout the entire ectodomain. In conclusion, our study reveals that the isotype of IgG does not necessarily determine the epitopes and pathogenicity of pemphigus autoantibodies. Pemphigus foliaceus (PF) is an antibody-mediated autoimmune disorder with IgG1 and IgG4 as the predominant subclasses of autoantibodies against a desmosomal glycoprotein, desmoglein-1 (Dsg1). Previously, we found that the IgG4 anti-Dsg1 autoantibodies only recognize a conformational epitope(s), whereas the IgG1 autoantibodies recognize both conformational and linear epitopes but do not display pathogenicity in the passive transfer animal model. The purpose of this study was to analyze the epitopes recognized by autoanti-bodies from a subset of PF patients who only express anti-Dsg1 of the IgG1 isotype throughout the course of their diseases and to further characterize the pathogenicity of their IgG1 anti-Dsg1. We found that IgG1 auto-antibodies in this subset of PF patients, similar to IgG4 autoantibodies from other PF patients, are able to bind both human and mouse skin and induce the experimental PF in mice. Moreover, a detailed epitope mapping reveals that the conformational epitopes recognized by IgG1 autoantibodies from these PF patients are restricted to the first 161 amino acids of Dsg1, whereas the linear epitopes are spread throughout the entire ectodomain. In conclusion, our study reveals that the isotype of IgG does not necessarily determine the epitopes and pathogenicity of pemphigus autoantibodies. desmoglein-1 desmoglein-3 glutathione S-transferase horseradish peroxidase immunoblotting immunofluorescence pemphigus foliaceus recombinant Dsg1 recombinant Dsg3 Pemphigus foliaceus (PF) is an autoimmune cutaneous disorder characterized by subcorneal blisters in the epidermis and circulating autoantibodies that recognize the desmosomal glycoprotein desmoglein-1 (Dsg1) (Civatte, 1943Civatte A. Diagnostic histopathologique de la dermatite polymorphe douloureseou maladie de During-Brocq.Ann Dermatol Syph. 1943; 3: 1-30Google Scholar;Lever, 1965Lever W.F. Pemphigus vulgaris.Pemphigus and Pemphigoid. C.C. Thomas, Springfield1965Google Scholar;Wheeler et al., 1991Wheeler G.N. Parker A.E. Thomas C.L. et al.Desmosomal glycoprotein DGI, a component of intercellular desmosome junctions, is related to the cadherin family of cell adhesion molecules.Proc Natl Acad Sci USA. 1991; 88: 4796-4800Crossref PubMed Scopus (152) Google Scholar;Buxton and Magee, 1992Buxton R.S. Magee A.I. Structure and interactions of desmosomal and other cadherins.Cell Biol. 1992; 3: 157-167Google Scholar). The target antigen, Dsg1, is a 160- kDa transmembrane protein and a member of the desmoglein subfamily of the cell adhesion supergene family (Wheeler et al., 1991Wheeler G.N. Parker A.E. Thomas C.L. et al.Desmosomal glycoprotein DGI, a component of intercellular desmosome junctions, is related to the cadherin family of cell adhesion molecules.Proc Natl Acad Sci USA. 1991; 88: 4796-4800Crossref PubMed Scopus (152) Google Scholar;Buxton and Magee, 1992Buxton R.S. Magee A.I. Structure and interactions of desmosomal and other cadherins.Cell Biol. 1992; 3: 157-167Google Scholar). The intracellular domain of Dsg1 interacts with a desmosomal plaque component, plakoglobin, linking it to the cytoskeleton (Korman et al., 1989Korman N.J. Eyre R.W. Klaus-Kovtun V. Stanley J.R. Demonstration of an adhering-junction molecule (plakoglobin) in the autoantigens of pemphigus foliaceus and pemphigus vulgaris.N Engl J Med. 1989; 321: 631-635Crossref PubMed Scopus (214) Google Scholar), whereas its extracellular domain harbors disease-relevant epitopes and might mediate the heterophilic interaction that brings about cell adhesion (Marcozzi et al., 1998Marcozzi C. Burdett I.D. Buxton R.S. Magee A.I. Coexpression of both types of desmosomal cadherin and plakoglobin confers strong intercellular adhesion.J Cell Sci. 1998; 111: 495-509PubMed Google Scholar). The subcorneal blisters in PF are formed by a process of keratinocyte cell–cell detachment known as acantholysis. In active PF, the Nikolsky's sign is easily elicited (Lever, 1965Lever W.F. Pemphigus vulgaris.Pemphigus and Pemphigoid. C.C. Thomas, Springfield1965Google Scholar). Unlike pemphigus vulgaris, mucosal blisters or erosions are not observed in PF patients. Recently, however, it was demonstrated by electron microscopy that several mucosal tissues exhibit signs of intercellular space widening, suggesting the possible involvement of mucosae in PF (Guedes et al., 2002Guedes A.C. Rotta O. Leite H.V. Leite V.H. Ultrastructural aspects of mucosas in endemic pemphigus foliaceus.Arch Dermatol. 2002; 138: 949-954Crossref PubMed Scopus (7) Google Scholar). In pemphigus vulgaris, approximately 50% of patients carry IgG autoantibodies against Dsg1 in addition to anti-desmoglein-3 (Dsg3), whereas in PF, only 7% of patients show autoantibody activity against Dsg3 (Arteaga et al., 2002Arteaga L.A. Prisayanh P.S. Warren S.J.P. Liu Z. Diaz L.A. Lin M.S. A subset of pemphigus foliaceus patients exhibits pathogenic autoantibodies against both desmogleins-1 and desmogleins-3.J Invest Dermatol. 2002; 118: 806-811Crossref PubMed Scopus (58) Google Scholar). Further autoantibody profile analysis reveals that the predominant isotypes of anti-Dsg1 in PF patients are IgG4 and IgG1 (Jones et al., 1988Jones C.C. Hamilton R.G. Jordon R.E. Subclass distribution of human IgG autoantibodies in pemphigus.J Clin Immunol. 1988; 8: 43-49Crossref PubMed Scopus (82) Google Scholar;Futei et al., 2001Futei Y. Amagai M. Ishii K. Kuroda-Kinoshita K. Ohya K. Nishikawa T. Predominant IgG4 subclass in autoantibodies of pemphigus vulgaris and foliaceus.J Dermatol Sci. 2001; 26: 55-61Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar;Warren et al., 2003Warren S.J.P. Arteaga L.A. Rivitti E.A. et al.The role of subclass switching in the pathogenesis of endemic pemphigus foliaceus.J Invest Dermatol. 2003; 120: 104-108Crossref PubMed Scopus (22) Google Scholar), with the titer of IgG4 closely associated with the activity of the disease, while the IgG1 titer remains constant throughout the course of the disease (Warren et al., 2003Warren S.J.P. Arteaga L.A. Rivitti E.A. et al.The role of subclass switching in the pathogenesis of endemic pemphigus foliaceus.J Invest Dermatol. 2003; 120: 104-108Crossref PubMed Scopus (22) Google Scholar). When evaluating the pathogenic activity of anti-Dsg1 subclass autoantibodies using a passive transfer animal model, it was demonstrated that only the IgG4, but not IgG1, autoantibodies were capable of inducing experimental PF (Rock et al., 1989Rock B. Martin C.R. Theofilopoulos A.N. et al.The pathogenic effect of IgG4 autoantibodies in endemic pemphigus foliaceus (endemic pemphigus foliaceus).N Engl J Med. 1989; 320: 1463-1469Crossref PubMed Scopus (233) Google Scholar). This result led to a belief that the inability of IgG1 anti-Dsg1 to induce the experimental PF is due to the lower titer or nonpathogenic activity of this subclass of autoantibodies or utilization of insufficient amounts of antibodies in the experiments. In light of our recent findings that contradict the common belief, it is now clear that the inability of IgG1 anti-Dsg1 to induce experimental PF in neonatal mice is because this subclass of autoantibodies, although it binds tissues from human skin and monkey esophagus, fails to cross-react with murine epidermis (Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar). Furthermore, we found that anti-Dsg1 of the IgG4 recognizes a conformational epitope(s), while the IgG1 anti-Dsg1 in PF patients recognizes both linear and conformational epitopes. In this study, we identified a unique subset of PF patients (4 of 52) who only express IgG1 autoantibodies by direct and indirect immunofluorescence (IF) staining and ELISA. Similar to IgG4 autoantibodies from other PF patients, the IgG1 autoantibodies from this subset of patients are able to reproduce the clinical, immunologic, and histologic features of PF when injected into neonatal mice. Using recombinant Dsg1/-3 domain-swapped proteins and competition ELISA, we further determined that the conformational epitopes recognized by these IgG1 autoantibodies are located in the first 87 amino acids of Dsg1, whereas the linear epitopes are spread throughout the ectodomain of Dsg1. The clinical and immunologic characterization of this group of patients further elucidates the complexity of autoimmunity in PF. Both human and animal protocols in this study have been approved by the MCW and in adherence to the Helsinki Principles. For human subject studies, we have obtained Exemption Four Status in our approval in which, we do not have direct contact with patients and are only allowed to use samples that will otherwise be discarded. Serum samples from 52 PF patients with active diseases were obtained from the dermatology clinic at the Medical College of Wisconsin. The diagnosis of PF was made on the basis of clinical examination and biopsy as well as direct and indirect IF staining. This study was approved by the Institutional Review Board of the Medical College of Wisconsin. Indirect IF was used to study the binding of the patients' autoantibodies to epidermal tissues from different sources (Beutner and Jordon, 1964Beutner E.H. Jordon R.E. Demonstration of skin antibodies in sera of patients with pemphigus vulgaris by indirect immunofluorescence staining.Proc Soc Exp Biol Med. 1964; 306: 505-510Crossref Scopus (484) Google Scholar). In this study, monkey esophagus, adult human skin from the thigh, and the dorsal skin of neonatal BALB/c mice were used as substrates. For Ig subclass staining, mouse anti-human IgG, IgG1, IgG2, IgG3, and IgG4 (Zymed Laboratories, San Francisco, CA) were used as secondary antibodies. The intercellular space staining pattern on the epidermal substrates was visualized by sheep anti-mouse IgG conjugated with fluorescein (Zymed). Subclass direct IF was performed by incubating perilesional skin from PF patients with mouse anti-human IgG1, IgG2, IgG3, or IgG4 and sheep anti-mouse IgG conjugated with fluorescein. The direct and indirect IF results were documented using a SPOT digital camera (Diagnostic Instruments, Inc., Sterling Heights, MI). Recombinant proteins encompassing the extracellular domain of Dsg1 (rDsg1) and Dsg3 (rDsg3) were prepared from a baculovirus expression system with a stretch of six histidines engineered immediately downstream of the recombinant proteins (Ding et al., 1999Ding X. Diaz L.A. Fairley J.A. Giudice G.J. Liu Z. The anti-desmoglein 1 autoantibodies in pemphigus vulgaris sera are pathogenic.J Invest Dermatol. 1999; 112: 739-743Crossref PubMed Scopus (109) Google Scholar). The domain-swapped proteins, Dsg31−161/Dsg1164−496, Dsg31−403/Dsg1404−496, Dsg11−161/Dsg3163−566, Dsg11−401/Dsg3405−566, Dsg11−87/Dsg387−566, and Dsg31−88/Dsg189−496 were constructed as described (Sekiguchi et al., 2001Sekiguchi M. Futei Y. Fujii Y. Iwasaki T. Nishikawa T. Amagai M. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins.J Immunol. 2001; 167: 5439-5448Crossref PubMed Scopus (142) Google Scholar). These proteins were purified by Ni-NTA column according the protocol provided by the manufacturer (Qiagen, Chatsworth, CA). Fragments spanning portions of Dsg1 were constructed as glutathione S-transferase (GST) fusion proteins and expressed in the prokaryotic system as described (Liu et al., 1995Liu Z. Diaz L.A. Swartz S.J. Troy J.L. Fairley J.A. Giudice G.J. Molecular mapping of a pathogenically relevant BP180 epitope associated with experimentally induced murine bullous pemphigoid.J Immunol. 1995; 155: 5449-5454PubMed Google Scholar). In total, 11 fusion proteins were produced and used in the study. The amino acid sequences of these proteins are Dsg1.1, AA 1–60; Dsg1.2, AA 54–93; Dsg1.3, AA 80–156; Dsg1.4, AA 137–199; Dsg1.5, AA 178–233; Dsg1.6, AA 213–278; Dsg1.7, AA 264–309; Dsg1.8, AA 303–344; Dsg1.9, AA 337–401; Dsg1.10, AA 381–444; and Dsg1.11, AA 424–501. Circulating Dsg1 autoantibodies were identified by Dsg1 ELISA. The Dsg1 subclass ELISAs were established according to previous publications except mouse anti-human IgG, IgG1, IgG2, IgG3, or IgG4 conjugated with horseradish peroxidase (HRP) (Zymed) were used as the secondary antibodies in this study (Arteaga et al., 2002Arteaga L.A. Prisayanh P.S. Warren S.J.P. Liu Z. Diaz L.A. Lin M.S. A subset of pemphigus foliaceus patients exhibits pathogenic autoantibodies against both desmogleins-1 and desmogleins-3.J Invest Dermatol. 2002; 118: 806-811Crossref PubMed Scopus (58) Google Scholar;Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar). To identify the conformational epitopes recognized by PF patients expressing only anti-Dsg1 of the IgG1 isotype, a competition ELISA was employed as previously described (Sekiguchi et al., 2001Sekiguchi M. Futei Y. Fujii Y. Iwasaki T. Nishikawa T. Amagai M. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins.J Immunol. 2001; 167: 5439-5448Crossref PubMed Scopus (142) Google Scholar). Briefly, serum samples from PF patients were first preincubated with various concentrations of Dsg-swapped proteins at room temperature for 1 h. Subsequently, the activity of anti-Dsg1 was assessed by ELISA. IB was performed as previously described (Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar). Briefly, 200 ng of rDsg1, rDsg3, and GST fusion proteins was resolved by SDS-PAGE, transferred to nitrocellulose (Gibco BRL, Gaithersburg, MD), and incubated with the patients' sera at a dilution of 1:100. Secondary antibody, mouse anti-human IgG, IgG1, IgG2, IgG3, and IgG4 conjugated with HRP (Zymed) was used at a dilution of 1:1000. After a 15-h incubation with the primary antibody at room temperature, blots were washed and then incubated for 2 h with HRP–anti-human IgG antibodies. The anti-Dsg1 signal was visualized by an enhanced chemiluminescence system (Amersham Biosciences, Piscataway, NJ) according to the manufacturer's protocol. To assess the pathogenic activity of the PF patients' autoantibodies, the passive transfer model was utilized (Arteaga et al., 2002Arteaga L.A. Prisayanh P.S. Warren S.J.P. Liu Z. Diaz L.A. Lin M.S. A subset of pemphigus foliaceus patients exhibits pathogenic autoantibodies against both desmogleins-1 and desmogleins-3.J Invest Dermatol. 2002; 118: 806-811Crossref PubMed Scopus (58) Google Scholar). Briefly, 100 μL of IgG (protein G column purified) from two PF patients was administered to neonatal BALB/c mice (1 mg/g body weight) by the intradermal route. In the preabsorption experiments, autoantibodies from PF patients were first incubated with equal quantity of rDsg1 or rDsg3 at room temperature for 1 h before injecting to neonatal BALB/c mice. The skins of the neonatal mice were examined 18 h after the IgG injection, and signs of cutaneous disease were recorded. The animals were then euthanized and skin specimens were obtained from lesional sites for histologic examination by hematoxylin and eosin staining and light microscopy. Fluorescein isothiocyanate-conjugated goat anti-human IgG was obtained commercially (Kirkegaard & Perry Laboratories, Gaithersburg, MD) and used to identify the deposits of PF IgG in the mouse skin by direct IF. To analyze the isotype profile of the PF auto-antibodies, ELISA and indirect IF microscopy were performed. In a total of 52 PF patients with active disease, we found 4 who expressed only anti-Dsg1 autoantibodies of the IgG1 isotype (not shown). None of the 4 patients tested were positive for the presence of anti-Dsg3 autoantibodies (not shown). By direct IF, only IgG1 was detected in the perilesional skins from patients' biopsies (not shown). There is no remarkable difference in the clinical expression of this subset of PF patients compared to other PF patients seen in our clinics. In a series of samples collected from these 4 patients over time, all tested positive only for the presence of IgG1 anti-Dsg1 by ELISA (not shown). To analyze the tissue specificity of autoantibodies from these patients, indirect IF was performed using epithelial tissues from monkey esophagus, adult human skin, and neonatal mouse skin. As shown in Figure 1, only total IgG (Figure 1a,d,g) and IgG1 (Figure 1b,e,h), but not IgG4 (Figure 1c,f,i) and other isotypes (not shown) of autoantibodies from these patients were able to stain these three substrates. These results were concordant with the finding that only IgG1, but no IgG2, IgG3, or IgG4, autoantibody deposits were detected in the perilesional skin of these patients by direct IF microscopy. Because IgG1 is the only isotype of anti-Dsg1 autoantibodies in these PF patients and it binds mouse skin (Figure 1h), it was of interest to know whether the IgG1 can elicit experimental PF. To assess the pathogenic activity of IgG1 autoantibodies in vivo, a mouse passive transfer model was utilized as described (Arteaga et al., 2002Arteaga L.A. Prisayanh P.S. Warren S.J.P. Liu Z. Diaz L.A. Lin M.S. A subset of pemphigus foliaceus patients exhibits pathogenic autoantibodies against both desmogleins-1 and desmogleins-3.J Invest Dermatol. 2002; 118: 806-811Crossref PubMed Scopus (58) Google Scholar). Because we only have enough serum samples from two of these four PF patients, passive transfer experiments were performed using affinity-purified autoantibodies from their samples and used to assess the pathogenic activities of their IgG1 autoantibodies. Both patients' autoantibodies elicited disease in neonatal mice. As demonstrated in Figure 2(a), IgG autoantibodies from a representative patient successfully induced skin lesions in neonatal mice with clinical, histologic, and immunologic features similar to those observed in human patients with PF, indicating that IgG1 anti-Dsg1 autoantibodies in this subset of patients are as effective in inducing experimental PF as IgG4 autoantibodies from other PF patients (Rock et al., 1989Rock B. Martin C.R. Theofilopoulos A.N. et al.The pathogenic effect of IgG4 autoantibodies in endemic pemphigus foliaceus (endemic pemphigus foliaceus).N Engl J Med. 1989; 320: 1463-1469Crossref PubMed Scopus (233) Google Scholar). The autoantibody-induced skin lesions in experimental PF were abolished by preabsorbing patients' IgG with equal quantity of rDsg1 (Figure 2a), but not rDsg3 (Figure 2c), further confirming the fact that IgG1 anti-Dsg1 autoantibodies caused superficial blisters in these patients. To analyze the conformational epitope(s) recognized by the IgG1 anti-Dsg1 autoantibodies from these patients, rDsg1, rDsg3, and domain-swapped Dsg proteins were used in the competition Dsg1 ELISA (Sekiguchi et al., 2001Sekiguchi M. Futei Y. Fujii Y. Iwasaki T. Nishikawa T. Amagai M. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins.J Immunol. 2001; 167: 5439-5448Crossref PubMed Scopus (142) Google Scholar). As shown in Figure 3(a), rDsg1, but not rDsg3, inhibited the activity of anti-Dsg1 in a concentration-dependent manner. To further determine the location of epitope(s) recognized by IgG1 anti-Dsg1 autoantibodies, domain-swapped Dsg proteins were used in the competition ELISA. As shown in Figure 3(b), the anti-Dsg1 activity of a representative PF patient was blocked by domain-swapped proteins Dsg11−161/Dsg3163−566 and Dsg11−401/Dsg3405−566 in a concentration-dependent manner, but not by Dsg31−161/Dsg1164−496 or Dsg31−403/Dsg1404−496. At a concentration of 5 μg/mL, both Dsg11−161/Dsg3163−566 and Dsg11−401/Dsg3405−566 were able to inhibit the activity of IgG1 anti-Dsg1 by 100% in the Dsg1 ELISA. Because the antigen-binding activities of anti-Dsg1 autoantibodies were inhibited by Dsg11−161/Dsg3163−566 and Dsg11−401/Dsg3405−566, but not by Dsg31−161/Dsg1164−496, our results suggested that the Dsg1 epitope(s) recognized by IgG1 anti-Dsg1 autoantibodies from this subset of PF patients is located on the first 161 amino acids of the Dsg1. To further investigate whether the conformational epitopes recognized by IgG anti-Dsg1 autoantibodies are located at the N-terminus of Dsg1, Dsg11−87/Dsg387−566 and Dsg31−88/Dsg189−496 were used in the competition Dsg1 ELISA. As demonstrated in Figure 3(c), the activity of anti-Dsg1 autoantibodies was blocked by Dsg11−87/Dsg387−566 (but not by Dsg31−88/Dsg189−496), following a concentration-dependent pattern, supporting the fact that the conformational epitopes recognized by IgG1 anti-Dsg1 auto-antibodies are located at the N-terminus of Dsg1. Previously, it was shown that IgG1 anti-Dsg1 autoantibodies from PF patients also recognize a linear epitope(s) (Emery et al., 1995Emery D.J. Diaz L.A. Fairley J.A. Lopez A. Taylor A.F. Giudice G.J. Pemphigus foliaceus and pemphigus vulgaris autoantibodies react with the extracellular domain of desmogleins-1.J Invest Dermatol. 1995; 104: 323-328Crossref PubMed Scopus (103) Google Scholar;Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar). To determine whether these four PF patients also had anti-Dsg1 that recognized the linear Dsg1 epitope(s), IB was performed. As demonstrated in Figure 4(a), autoantibodies from these PF patients only recognized rDsg1, but not rDsg3, by IB, suggesting that anti-Dsg1 from this subset of PF patients also bound the linear epitope(s) on the ectodomain of Dsg1. There were no other IgG subclass autoantibodies that recognized rDsg1 by IB except IgG1 (Figure 4b); these results further confirmed that IgG1 was the sole isotype of anti-Dsg1 in these patients. To further identify whether anti-Dsg1 autoantibodies against linear Dsg1 epitopes are independent from those recognizing conformational epitopes, IB against rDsg1 was performed using serum preabsorbed with either rDsg1 or rDsg3. As shown in Figure 4c, serum samples at a dilution of 1:100 preabsorbed with 5 μg of rDsg1, although lost activity to rDsg1 in ELISA (Figure 3), continued to react with denatured rDsg1 in IB. These results suggested that autoantibodies recognizing linear Dsg1 epitopes are distinct from those binding to conformational epitopes. Upon identification of the linear Dsg1 epitope(s), we found that serum from these patients bound fusion proteins spread throughout the extracellular domain of Dsg1 (Table I; Figure 4d). No common linear epitopes among these four PF patients could be identified by the IB studies. Along with the indirect IF and ELISA results, these findings suggest that there were two populations of IgG1 anti-Dsg1 autoantibodies in this unique subset of PF patients, one recognizing a conformational, while the other binding to the linear epitope(s) on the ectodomain of Dsg1.Table IIgG1 anti-Dsg1 autoantibodies recognize linear epitopes on the ectodomain of Dsg1aTo determine the linear epitope(s) recognized by anti-Dsg1 from PF patients expressing only IgG1 autoantibodies, IB assay was performed as described under Materials and Methods. A Dsg1 fragment that was bound by autoantibodies was recognized as harboring an anti-Dsg1 epitope(s) and marked as "+"; a fragment that was not bound by autoantibodies was considered negative for anti-Dsg1 epitope and marked as "(–)."IgG1 anti-Dsg1 autoantibodyPatient1.11.21.31.41.51.61.71.81.91.101.11PF1+(–)(–)(–)++(–)(–)+(–)+PF2(–)(–)(–)(–)(–)(–)(–)+(–)+(–)PF3(–)+(–)(–)(–)(–)(–)(–)(–)(–)+PF4(–)(–)+(–)(–)(–)(–)(–)(–)+(–)a To determine the linear epitope(s) recognized by anti-Dsg1 from PF patients expressing only IgG1 autoantibodies, IB assay was performed as described under Materials and Methods. A Dsg1 fragment that was bound by autoantibodies was recognized as harboring an anti-Dsg1 epitope(s) and marked as "+"; a fragment that was not bound by autoantibodies was considered negative for anti-Dsg1 epitope and marked as "(–)." Open table in a new tab IgG1 and IgG4 are the two predominant isotypes of autoantibodies in patients with pemphigus; nevertheless, the clinical and immunologic implications of the dual isotypes of anti-Dsg are not yet clear. In pemphigus vulgaris, distinct epitopes are recognized by these two isotypes of anti-Dsg3 (Bhol et al., 1995Bhol K. Natarajan K. Nagarwalla N. Mohimen A. Aoki V. Ahmed A.R. Correlation of peptide specificity and IgG subclass with pathogenic and nonpathogenic autoantibodies in pemphigus vulgaris: A model for autoimmunity.Proc Natl Acad Sci USA. 1995; 92: 5239-5243Crossref PubMed Scopus (166) Google Scholar). It was suggested that IgG4 anti-Dsg3 is directly linked to the activity of the disease (Bhol et al., 1994Bhol K. Mohimen A. Ahmed A.R. Correlation of subclasses of IgG with disease activity in pemphigus vulgaris.Dermatol. 1994; 189: 85-89Crossref PubMed Scopus (71) Google Scholar). In a recent study of the endemic PF in Brazil, it was also demonstrated that IgG1 anti-Dsg1 is associated with the remission of the disease, whereas the IgG4 isotype of autoantibodies is closely related to the activity of PF (Warren et al., 2003Warren S.J.P. Arteaga L.A. Rivitti E.A. et al.The role of subclass switching in the pathogenesis of endemic pemphigus foliaceus.J Invest Dermatol. 2003; 120: 104-108Crossref PubMed Scopus (22) Google Scholar). Although, IgG1 anti-Dsg1 autoantibodies could be readily identified in endemic PF patients before the onset of their disease (Lin et al., 2001Lin M.S. Arteaga L.A. Prisayanh P.S. Warren S.J.P. Liu Z. Diaz L.A. Isotype and epitope profiles of anti-desmoglein-1 autoantibodies in patients and normal subjects in an endemic focus of fogo selvagem.J Invest Dermatol. 2001; 117: 460Google Scholar), the pathophysiologic role of IgG1 anti-Dsg1 autoantibodies in these individuals is still elusive. From this previous study, it is clear that IgG1 anti-Dsg1 autoantibodies from normal individuals recognize denatured rDsg1 by IB and bind tissues only from monkey esophagus but not tissues from human or murine skin by indirect IF; nevertheless, the epitope specificity of these autoantibodies is not clear. Moreover, purified anti-Dsg1 autoantibodies from a normal individual failed to induce disease in the passive transfer experiments, suggesting that IgG1 anti-Dsg1 from normal individuals in the endemic PF areas may not play a direct role in the development of the disease. Recently, we reported that IgG1 and IgG4 anti-Dsg1 in PF patients demonstrated distinct tissue and epitope specificity (Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar). It appears that anti-Dsg1 autoantibodies of the IgG1 isotype recognize both linear and conformational epitope(s) and fail to bind mouse skin, while its IgG4 counterpart recognizes only a conformational epitope(s) but can bind mouse skin. This finding suggests that the inability of IgG1 anti-Dsg1 to elicit the experimental PF in neonatal mice is simply because this isotype of autoantibodies in some PF patients does not cross-react with mouse skin, but not due to a lower pathogenicity of the autoantibodies of this isotype or an insufficient amount of autoantibodies used in previous studies (Rock et al., 1989Rock B. Martin C.R. Theofilopoulos A.N. et al.The pathogenic effect of IgG4 autoantibodies in endemic pemphigus foliaceus (endemic pemphigus foliaceus).N Engl J Med. 1989; 320: 1463-1469Crossref PubMed Scopus (233) Google Scholar). In this report, we identified a unique subset of PF patients who only express IgG1 anti-Dsg1 autoantibodies throughout the course of their disease. This conclusion was drawn based on the observation that both direct and indirect IF microscopy and Dsg1 ELISA only detect IgG1 anti-Dsg1 in samples collected from these four patients. Our findings are clearly different from those reported previously, which demonstrated that IgG4 is the predominant isotype of anti-Dsg1 in patients with PF (Futei et al., 2001Futei Y. Amagai M. Ishii K. Kuroda-Kinoshita K. Ohya K. Nishikawa T. Predominant IgG4 subclass in autoantibodies of pemphigus vulgaris and foliaceus.J Dermatol Sci. 2001; 26: 55-61Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar;Warren et al., 2003Warren S.J.P. Arteaga L.A. Rivitti E.A. et al.The role of subclass switching in the pathogenesis of endemic pemphigus foliaceus.J Invest Dermatol. 2003; 120: 104-108Crossref PubMed Scopus (22) Google Scholar). Interestingly, this subset of patients does not express distinct clinical features as those PF patients producing both IgG1 and IgG4 anti-Dsg1 autoantibodies. Assessing the pathogenicity of autoantibodies using a passive transfer animal model, we demonstrated that the IgG1 anti-Dsg1 autoantibodies from these patients were capable of eliciting PF-like lesions in neonatal mice (Figure 2), providing the first evidence that this subclass of anti-Dsg1 is able to induce experimental PF in the passive transfer model. Comparing the clinical, histologic, and immunologic features in the experimental PF mice induced by IgG1 anti-Dsg1 autoantibodies, there is no noticeable difference from those elicited by IgG4 autoantibodies from patients with endemic PF (Rock et al., 1989Rock B. Martin C.R. Theofilopoulos A.N. et al.The pathogenic effect of IgG4 autoantibodies in endemic pemphigus foliaceus (endemic pemphigus foliaceus).N Engl J Med. 1989; 320: 1463-1469Crossref PubMed Scopus (233) Google Scholar). Because IgG1 and IgG4 anti-Dsg1 autoantibodies from patients with PF exhibit distinct tissue and antigen-binding specificity (Hacker et al., 2002Hacker M.K. Jansen M. Fairley J.A. Lin M.S. Isotype and antigenic profiles of pemphigus foliaceus and pemphigus vulgaris autoantibodies.Clin Immunol. 2002; 105: 64-74Crossref PubMed Scopus (48) Google Scholar), we speculate that anti-Dsg1 from these IgG1 PF patients may recognize unique Dsg1 epitope(s). Nevertheless, from the competition ELISA using domain-swapped Dsg proteins, we found that IgG1 autoantibodies from these patients appear to recognize similar if not the same conformational epitope(s) (i.e., the first 87 amino acids of Dsg1 ectodomain) (Figure 3) as those of other PF patients (Sekiguchi et al., 2001Sekiguchi M. Futei Y. Fujii Y. Iwasaki T. Nishikawa T. Amagai M. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins.J Immunol. 2001; 167: 5439-5448Crossref PubMed Scopus (142) Google Scholar). This finding along with the previous report supports the conclusion that the N-terminus of Dsg1 harbors key epitope(s) that are critical for the development of PF (Amagai et al., 1995Amagai M. Hashimoto T. Green K.J. Shimizu N. Nishikawa T. Antigen-specific immunoabsorption of pathogenic autoantibodies in pemphigus foliaceus.J Invest Dermatol. 1995; 104: 895-901Crossref PubMed Scopus (225) Google Scholar;Kowalczyk et al., 1995Kowalczyk A.P. Anderson J.E. Borgwardt J.E. Hashimoto T. Stanley J.R. Green K.J. Pemphigus sera recognize conformationally sensitive epitopes in the amino-terminal region of desmogleins-1.J Invest Dermatol. 1995; 105: 147-152Crossref PubMed Scopus (74) Google Scholar;Sekiguchi et al., 2001Sekiguchi M. Futei Y. Fujii Y. Iwasaki T. Nishikawa T. Amagai M. Dominant autoimmune epitopes recognized by pemphigus antibodies map to the N-terminal adhesive region of desmogleins.J Immunol. 2001; 167: 5439-5448Crossref PubMed Scopus (142) Google Scholar). In this subset of PF patients, we also found that the IgG1 anti-Dsg1 autoantibodies reacted with linear epitopes (Figure 4a,b). Interestingly, serum samples from these patients maintained activity to denatured rDsg1 after preabsorption with rDsg1 in Ca2+-abundant buffer (Figure 4c), suggesting that there are two populations of autoantibodies in these PF patients, one recognizing the conformational epitopes and the other binding the linear epitopes. A detailed epitope mapping study revealed that the linear Dsg1 epitopes were not limited to the N-terminus of Dsg1 (Figure 4d). It appears that the linear epitopes recognized by IgG anti-Dsg1 of these patients are spread out along the ectodomain of Dsg1 (Table I). From the four PF patients tested here, along with a previous report (Arteaga et al., 2000Arteaga L. Warren S. Liu Z. Lin M. Diaz L. Desmoglein-1 epitopes recognized by autoantibodies from patients with fogo selvagem.J Invest Dermatol. 2000; 114: 804Google Scholar), no common Dsg1 linear epitopes recognized by PF autoantibodies could be identified. When the binding activities of anti-Dsg1 autoantibodies to Dsg1 fusion proteins were removed by preabsorption, the serum samples continued to react with Dsg1 by ELISA and indirect IF (not shown). It is possible that the production of autoantibodies to linear Dsg1 epitopes in PF patients is an epi-phenomenon of autoimmune responses developed from the destruction of epidermis induced by pathogenic anti-Dsg1 and the subsequent exposure of epitopes on the ectodomain of Dsg1 that are inaccessible to the autoimmune reaction. The relationship of these linear epitopes to the development of disease is unknown. In conclusion, we found that IgG1 anti-Dsg1 autoantibodies from this particular subset of PF patients share similar properties to those of IgG4 anti-Dsg1 autoantibodies from other PF patients. This includes the induction of experimental PF in the passive transfer animal model and binding both human and mouse skin. Unlike IgG4 anti-Dsg1, which does not bind the denatured form of Dsg1, IgG1 autoantibodies from these patients did react with the Dsg1 ectodomain by IB. At present, it is not clear what mechanism(s) governs the production of only IgG1, but not IgG4, autoantibodies that recognize the linear epitopes on Dsg1 in PF patients. Our identification of this unique subset of PF patients expressing only pathogenic IgG1 autoantibodies throughout the course of their disease further illustrates the complexity of autoimmune response in PF. Furthermore, the demonstration of the pathogenic IgG1 antibodies in PF suggests that the IgG subclass does not necessarily determine the epitopes and pathogenicity of pemphigus autoantibodies. The authors thank Dr Zhi Liu at the Department of Dermatology, University of North Carolina at Chapel Hill, for his design and construction of Dsg1–GST fusion proteins and Dr Zelmira Lazarova for her invaluable suggestions. This study was supported in part by U.S. Public Health Service Grants RO1-AI48348 (M.S.L.) from the National Institutes of Health and by a Merit Award from the Veterans Administration Central Office (J.A.F).