Molecular Mapping of the Major Epitopes of BP180 Recognized by Herpes Gestationis Autoantibodies

Clinical Immunology Vol. 92, No. 3, September, pp. 285–292, 1999 Article ID clim.1999.4756, available online at http://www.idealibrary.com on

Molecular Mapping of the Major Epitopes of BP180 Recognized by Herpes Gestationis Autoantibodies Mong-Shang Lin,* Manish Gharia,* Chang-Ling Fu,* Monica Olague-Marchan,* ,† Mary Hacker,* ,† Karen E. Harman,‡ Balbir S. Bhogal,‡ Martin M. Black,‡ Luis A. Diaz,* ,§ and George J. Giudice* ,† *Department of Dermatology and †Department of Biochemistry, Medical College of Wisconsin, Milwaukee, Wisconsin 53226; §VA Medical Center, Milwaukee, Wisconsin; and ‡St. John’s Institute of Dermatology, St. Thomas’ Hospital, London SE1 7EH, United Kingdom

Herpes gestationis (HG) is an autoantibody-mediated subepidermal bullous dermatosis associated with pregnancy. The primary target of HG autoantibodies is BP180, a 180-kDa hemidesmosomal glycoprotein. We previously showed that autoantibodies and autoimmune T lymphocytes from HG patients recognize the MCW-1 antigenic site (AA 507–520), which is located in the membrane-proximal noncollagenous domain (NC16A) of BP180. Here, we analyzed the sera of 37 HG patients to further define the sites on BP180 that are targeted by autoantibodies. All of the HG sera, but none of the control sera, were immunoreactive with sec180e, a 120-kDa recombinant protein encompassing the entire BP180 extracellular domain. HG sera depleted of reactivity to NC16A no longer reacted with sec180e, indicating that the major HG-associated epitopes on BP180 are restricted to the NC16A domain. The vast majority of the HG sera (34 of 37) reacted with a 7 amino acid peptide corresponding to the N-terminal half of MCW-1 (MCW-1A). Eleven HG sera (including the 3 that failed to react with MCW-1A) recognized one or more of three antigenic sites located within a 15 amino acid stretch immediately downstream of MCW1A. In summary, we have identified four major HGassociated epitopes clustered within a 22 amino acid region of the BP180 ectodomain. These findings support the hypothesis that an autoimmune response to the BP180 NC16A domain is a crucial step in the pathogenesis of HG. © 1999 Academic Press Key Words: autoimmunity; keratinocyte; bullous disease; hemidesmosome. INTRODUCTION

Herpes gestationis (HG) is a nonherpetic autoimmune disease of women characterized by the development of subepidermal inflammatory blisters during the last trimester of pregnancy (1). Typically, perilesional skin of these patients shows deposition of C3 at the cutaneous basement membrane zone (BMZ) by direct immunofluorescence (IF) techniques. Although the sera of most (.70%) HG patients yield negative results

when assayed for BMZ-reactive autoantibodies by routine indirect IF procedures, the majority of these same sera exhibit complement-fixing anti-BMZ autoantibodies (termed “HG factor”) when assayed using a modified complement-indirect IF protocol (2– 4). HG antiBMZ autoantibodies belong to the IgG class and are predominantly of the IgG1 subclass (5, 6). HG autoantibodies recognize the ectodomain of BP180, a hemidesmosomal protein also known as BPAg2 and type XVII collagen. The BP180 antigen is also targeted by IgG autoantibodies produced by patients suffering from bullous pemphigoid (BP), an inflammatory subepidermal blistering disease most frequently observed in the elderly (7–9). Based on the similar clinical, histologic, and immunological findings in HG and BP, it has been hypothesized that these diseases share key elements in their immunopathogenic mechanisms. Interestingly, rabbits immunized with the murine homologue of BP180 produce anti-BP180 antibodies that are pathogenic when tested by passive transfer experiments in neonatal BALB/c mice (10). The disease in these animals, which duplicates the key features of HG and BP, was found to be dependent on complement activation and neutrophil recruitment to the site of subepidermal blister formation (11, 12). In addition, it is well documented that HG, a disease that affects approximately 1 in 100,000 pregnant women, is associated with a blistering disease in the newborn (1). In these rare cases, the neonatal blistering disease is similar to HG and BP, but is transient, disappearing within the first few weeks of life, suggesting a transplacental passage of pathogenic HG autoantibodies. Moreover, passive transfer of anti-BP180 antibodies to late-stage pregnant mice has produced subepidermal blisters in the neonatal offspring (Z. Liu and L. A. Diaz, unpublished observation), further supporting the above interpretation and substantiating the relevance of this model to the human disease. It appears, therefore, that anti-BP180 autoantibodies present in the sera of HG and BP patients play a

285

1521-6616/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

286

LIN ET AL.

key role in the pathogenesis of these diseases. We have previously shown that BP autoantibodies recognize epitopes clustered within the NC16A domain of the BP180 antigen (13–17). In the present study we have performed a rigorous epitope mapping analysis on a large number of HG patients’ sera and have shown that the NC16A domain of BP180 also harbors the major epitopes targeted by HG autoantibodies. MATERIALS AND METHODS

HG Patients and Controls In this study, sera were obtained from 37 HG patients. Clinically, these patients showed linear IgG and C3 deposits in the basement membrane zone of the lesional skin. Indirect immunofluorescence staining was performed using human foreskin as substrate to detect IgG anti-BMZ antibodies (17). Serum samples from patients with other cutaneous autoimmune diseases, such as pemphigus vulgaris (PV; n 5 4), pemphigus foliaceus (PF; n 5 4), systemic lupus erythematosus (n 5 2), and psoriasis (n 5 2), were included along with sera from normal individuals (n 5 8) as controls. Preparation of Recombinant Forms of BP180 The entire extracellular domain of BP180 was expressed as a secreted protein (designated sec180e) in cultured mammalian cells using the pCEP4 expression system (Invitrogen). The first step in the generation of this expression construct, pCEP4sec180e, involved insertion of the cDNA encoding the N-terminal signal and propeptide sequences of desmoglein-1 (nucleotides 63 to 236) into the HindIII and NotI sites of pCEP4. The encoded desmoglein-1 peptide was used to ensure that sec180e would be targeted to the secretory pathway. This stretch of the desmoglein-1 cDNA was obtained by PCR amplification using the pSVsec180e construct (18) as the template. In the second step, the cDNA segment encoding the extracellular domain of human BP180 (AA 490 –1497) was obtained as a NotI restriction fragment of pSVsec180e (18) and subcloned into the NotI site of pCEP4dsg1-leader. The final construct, pCEP4sec180e, was sequenced using the ALF DNA automated sequencer and the Autoread sequencing kit (Pharmacia-Biotech) with fluorescein-conjugated vector and BP180 primers to confirm the orientation, sequence, and registry of the desmoglein-1- and BP180-coding regions. pCEP4sec180e was transfected into the embryonic human kidney cell line, 293-EBNA (Invitrogen), using the following protocol. The 293-EBNA cells were incubated for 18 h with a mixture of DNA:lipofectin reagent (Gibco-BRL Life Technologies), after which the DNA

mixture was replaced with normal growth medium (DMEM with 10% fetal bovine serum). After an additional 24-h incubation, the cells were subcultured in the presence of the selection antibiotics hygromicin B (Invitrogen) and G418-sulfate (Calbiochem). The cells that survived selection were cloned out in 96-well plates by limiting dilution. Expression and secretion of sec180e were confirmed by immunoblot analysis of the culture conditioned medium using rabbit anti-BP180 antisera (19). One clone expressing a high level of the 120-kDa immunoreactive sec180e was expanded and the culture supernatant fraction from this clone was used in the immunoblot analysis of the HG patients’ sera. Epitope mapping experiments were performed using a panel of BP180 – glutathione S-transferase (GST) fusion proteins—NC16A (AA 490 –562), NC16A1 (AA 490 –506), NC16A1-3 (AA 490 –534), NC16A2-5 (AA 507– 630 and 830 – 855), NC16A2-4 (AA 507–548), NC16A2-3 (AA 507–534 and 836 – 855), NC16A2 (AA 507–520), NC16A2.5 (AA 514 –532), and NC16A3 (AA 521–534)—as shown in Fig. 1. The generation of these expression constructs has been documented in previous publications (13, 17, 19). These BP180 –GST fusion proteins were expressed in Escherichia coli strain DH5a and purified by glutathione agarose affinity chromatography (20). The purified fusion proteins were dialyzed against PBS, concentrated by ultrafiltration, and filter sterilized. The protein concentration was determined by the Bradford protein assays (BioRad, Hercules, CA). Immunoblot Analysis Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and immunoblotting were performed as described (17). Briefly, sec180e or fusion proteins were fractionated on 7.5 or 12% polyacrylamide gels, respectively, and transferred to nitrocellulose (Gibco-BRL). Blots were blocked with PTX buffer (0.01 M phosphate, pH 7.5, 0.2% Triton X-100, 0.15 M NaCl, 1 mM EGTA, 4% BSA) for 45 min. Sera from HG patients or controls were diluted 100-fold in PTX buffer and incubated with the blots for 15 h. The blots were washed, blocked for 30 min with PTX buffer, and incubated for 2 h with 125I-labeled Staphylococcus aureus protein A (10 5 cpm/ml). Subsequently, the blots were washed and the bound antibodies were visualized by autoradiography. Immunoadsorption Immunoadsorption procedures were conducted as described previously (17, 21). The sera were diluted in PTX and incubated overnight with recombinant GST or with GST fusion proteins containing segments of NC16A. The mixtures were then centrifuged as

AUTOANTIBODY EPITOPES IN HERPES GESTATIONIS

287

FIG. 1. Diagram of the recombinant BP180 ectodomain and fusion proteins encompassing the NC16A region. BP180 contains an amino-terminal cytoplasmic region and a carboxy-terminal extracellular domain, which are shown to the left and right, respectively, of the transmembrane domain (labeled “TM”). The ectodomain consists of 15 interrupted collagen domains depicted as filled rectangles. Sec180e, a recombinant protein encompassing the entire extracellular domain of BP180, is shown as the horizontal bar above BP180. Each of the NC16A fusion proteins shown below BP180 possesses an N-terminal GST moiety. Based on the immunoblotting results, HG sera included in this study can be categorized into four groups, as shown in the table to the right of this figure. 1 and 2, positive and negative immunoblotting results, respectively.

10,000g for 15 min at 4°C. The preabsorbed sera were subsequently used in the immunoblotting assays against various NC16A fusion proteins or sec180e. RESULTS

Sera of HG Patients React with the Ectodomain of BP180 For use in immunological assays, a recombinant form of the BP180 ectodomain, sec180e, was expressed as a soluble, secreted protein in the 293EBNA cell line. The culture supernatant fraction of sec180e-expressing cells was used in immunoblotting experiments. As shown in Fig. 2, sera from representative HG patients contain IgG autoantibodies that recognize sec180e (Fig. 2, lanes 1 and 3). These HG sera showed no specific immunoreactivity with components of the culture supernatant of vector-transfected control cells (Fig. 2, lanes 2 and 4). As summarized in Table 1, sera from all 37 HG patients contained IgG reactivity with sec180e. Sera from normal individuals (n 5 8) or from patients with unrelated cutaneous disorders, such as PV (n 5 4), PF (n 5 4), lupus erythematosus (n 5 2),

and psoriasis (n 5 2), did not react with sec180e. These results indicate that HG sera contain IgG autoantibodies directed against the BP180 ectodomain and that this reactivity appears to be specific for subepidermal bullous diseases, which includes BP and HG. Sera of HG Patients Recognize Epitopes within the NC16A Domain To further define the epitope(s) recognized by sera of HG patients, immunoblotting assays were performed using GST–NC16A fusion proteins as substrates. These data are reported in tabular form in Fig. 1, and representative results are shown in Fig. 3. All 37 HG sera reacted with fusion proteins NC16A1-5, NC16A1-3, NC16A2-5, NC16A2-4, and NC16A2-3, all five of which contain the BP180 segment AA 507–534. None of the 37 HG sera reacted with NC16A1. The C-terminal of NC16A (AA 535–562) was also found to be unreactive with HG sera based on immunoadsorption data. After depleting the HG sera of reactivity against NC16A1-3, they no longer recognized NC16A1-5 (data not shown). These data suggested that

288

LIN ET AL.

FIG. 2. Sera of HG patients specifically recognize sec180e recombinant proteins. Sec180e, a 120-kDa protein secreted by transfected 293EBNA cells, was fractioned by SDS–PAGE, transferred to nitrocellulose, and labeled with sera from HG patients (1:100 dilution). Lanes 1, 3, 5, and 7, loaded with sec180e; lanes 2, 4, 6, and 8, loaded with the supernatant fraction from untransfected 293EBNA cells. The primary antibodies used in the labeling reactions are as follows: lanes 1 and 2, serum of HG patient A; lanes 3 and 4, serum of HG patient B; lanes 5 and 6, serum from a normal control; lanes 7 and 8, rabbit anti-NC16A (1:1000).

major epitopes recognized by HG sera are located between AA 507 and 534. The HG sera could be categorized into four groups based on their reactivity profile with three overlapping BP180 fusion proteins that cover this central antigenic stretch of NC16A2 (containing the MCW-1 peptide), NC16A2.5, and NC16A3. The first and largest group of HG sera (Group I, composed of 26 of the 37 sera) reacted with NC16A2, but not with NC16A2.5 or NC16A3. The second group of HG sera (Group II, n 5 5) reacted with both NC16A2 and NC16A2.5. Group III (n 5 3) reacted with all three fusion proteins, and Group IV (n 5 3) reacted with NC16A2.5 and NC16A3 but not NC16A2 (see Fig. 1). To more precisely define the HG-associated epitopes on BP180, studies were carried out in which the sera were first depleted of reactivity against one NC16A peptide and then assayed by immunoblotting for reactivity with a second NC16A segment. These experimental results are reported in terms of the four groups of HG sera defined above. Preadsorption of Groups II and III sera with NC16A2.5 resulted in a diminution, but not elimination, of reactivity with NC16A2. Preadsorption of these same eight sera with NC16A2 reduced, but did not abolish, their immunoreactivity with NC16A2.5. The Group III sera preadsorbed against NC16A3 continued to react with NC16A2.5, although not as strongly as the unadsorbed or GSTadsorbed sera. Preadsorption of Group IV sera with NC16A3 knocked out their reactivity with NC16A2.5. After preadsorption of both Group III and IV sera with NC16A2.5, these six sera no longer reacted with NC16A3. These immunoadsorption results are presented in Table 2. HG Autoantibody Epitopes Are Restricted to the NC16A Region In order to determine whether regions of BP180 outside of the NC16A domain bear HG autoantibodyreactive sites, HG sera were depleted of reactivity to

NC16A and then assayed by immunoblotting for reactivity with sec180e. Figure 4 shows the immunoblotting results from representative HG sera. A Group I HG serum (which reacts exclusively with MCW-1) recognizes sec180e by immunoblotting (Fig. 4, lane 1). Preadsorption of this HG serum with NC16A2 (MCW-1) abolishes its reactivity with sec180e (Fig. 4, lane 2). This same pattern of reactivity was obtained with all HG sera in Groups I and II. A Group IV HG serum (which reacts with the overlapping BP180 peptides NC16A2.5 and NC16A3) also shows reactivity with sec180e (Fig. 4, lane 3). Preadsorption of this HG serum with NC16A3 abolishes its reactivity with sec180e (lane 4). Similar results were obtained with the other Group IV sera. In summary, using this liquid-phase preadsorption technique, we found that all 37 HG sera lost the ability to bind to the sec180e recombinant protein when immunoreactivity to NC16A was eliminated. These data indicate that there are no major HGassociated epitope(s) within the BP180 ectodomain outside of those defined in the NC16A region. DISCUSSION

This investigation clearly demonstrates that the vast majority of HG patients (37 of 37 in this study) have TABLE 1 Immunoreactivity of HG Patients to sec180e Recombinant Proteins Patients

IgG immunoreactivity to sec180e

HG PV PF SLE Psoriasis Normal individual

37/37 (100%) a 0/4 0/4 0/2 0/2 0/8

a Positive reactivity is reported as a fraction of those sera that reacted with sec180e.

289

AUTOANTIBODY EPITOPES IN HERPES GESTATIONIS

FIG. 3. HG autoantibodies recognize multiple sites within the BP180 NC16A domain. Sera of two HG patients were preadsorbed with recombinant GST and then assayed by immunoblotting for reactivity with the following fusion proteins: lane 1, GST; lane 2, NC16A2-5; lane 3, NC16A1-3; lane 4, NC16A2-4; lane 5, NC16A2-3; lane 6, NC16A1; lane 7, NC16A2; lane 8, NC16A3.

circulating IgG autoantibodies that react specifically with the extracellular domain of the BP180 antigen. This was demonstrated by immunoblot analysis using sec180e, a recombinant protein corresponding to the BP180 ectodomain generated in a mammalian expression system. The specificity of this reaction was established with the finding that sec180e was not recognized by any of the control sera, which included normal human sera and sera from patients with unrelated der-

matoses—pemphigus vulgaris, pemphigus foliaceus, psoriasis, and lupus erythematosus. Furthermore, we showed that anti-BP180 HG autoantibodies have highly restricted fine specificities. After depleting immunoreactivity to the major noncollagenous domain, NC16A, the HG sera no longer reacted with sec180e, indicating that this protein stretch contains the immunodominant autoantibody-reactive sites. One NC16A site, a 14 amino acid stretch (RSILPYGDSMDRIE)

TABLE 2 Epitope Mapping of HG Autoantibodies by Immunoadsorption Analysis Immunoblot reactivity with HG sera

Adsorbed with

NC16A2(MCW-1)

NC16A3

NC16A2.5

Group I (n 5 26) Group II (n 5 5)

NC16A2(MCW-1) NC16A2(MCW-1) NC16A3 NC16A2(MCW-1) NC16A2.5 NC16A3 NC16A2.5 NC16A3

2 2 1 2 111 1 2 2

2 2 2 111 2 2 2 2

2 1 2 1 1 2 2 2

Gropu III (n 5 3) Group IV (n 5 3)

Note. Liquid-phase immunoadsorption was utilized in the immunoblotting assays as described. HG sera preadsorbed with NC16A2, NC16A2.5, or NC16A3 were subsequently incubated with NC16A2, NC16A2.5, or NC16A3. The activities of HG IgG against these fusion proteins were documented by autoradiographs. 111, strong reactivity; 1, weak reactivity; 2, no reactivity.

290

LIN ET AL.

FIG. 4. HG IgG autoantibodies do not recognize epitopes outside of the NC16A region. HG serum samples recognized NC16A2 (MCW-1) (lanes 1 and 2) or NC16A3 (lanes 3 and 4) were preabsorbed with NC16A2 or NC16A3 fusion proteins, respectively. These preabsorbed sera were subsequently incubated with sec180e proteins (lanes 2 and 4). The IgG autoantibodies bound to the sec180e were further detected by 125 I-labeled protein A as described. Lane 5, rabbit anti-NC16A (1:1000).

designated MCW-1, was recognized by 34 of the 37 HG sera assayed, and in 26 of these cases (Group I HG sera characterized in Fig. 1), no other anti-BP180 autoantibodies were detectable other than those directed against MCW-1. The relevance of the MCW-1 antigenic site in the pathogenesis of HG is substantiated by our recent work on the characterization of the T cell response in HG patients. Multiple T cell clones derived from an HG patient showed specific reactivity with the MCW-1 peptide (22). These HG T cell clones expressed a Th1 cytokine profile, indicating that these cells might function in promoting the production of IgG1 anti-BP180 autoantibodies. There are other reports in the literature documenting the juxtaposition of epitopes that are reactive with helper T cells and antibodies. For example, T cell epitopes associated with experimental autoimmune myasthenia gravis were mapped to a peptide stretch of the acetylcholine receptor that encompasses residues implicated in a pathogenic B cell epitope (23). Specifically, these authors showed that a 17-mer peptide was capable of both activating helper T cells and stimulating the production of pathogenic autoantibodies in the experimental animals. Also, autoantibodies isolated from multiple sclerosis patients were shown to recognize a 10 amino acid peptide that was located within the epitope autoantigenic to the autoimmune T lymphocytes (24). In a similar vein, our findings indicate that the MCW-1 site on the extracellular domain of BP180 harbors epitopes targeted by both helper T cells and autoantibodies. Although the pathogenicity of the human autoantibodies has not been directly demonstrated, there is abundant evidence that a subepidermal blistering disease that mimics HG and BP can be induced in mice by the passive transfer of antiBP180 antibodies (10). And the pathogenically relevant epitope in this model system has been mapped to a site on murine BP180 that aligns with a site immediately downstream of MCW-1 on human BP180 (20).

Taken together, these findings provide a strong argument that MCW-1 plays a key role in triggering the autoimmune response in HG patients. A detailed epitope mapping study revealed that the MCW-1 site actually harbors at least two HG-associated epitopes and that the N-terminal half (a seven amino acid stretch) is recognized by all 34 of the MCW1-positive HG sera. This conclusion was drawn from two sets of data. First, the Group I HG sera were shown to react with MCW-1, but not with NC16A2.5, thus eliminating the region of overlap (i.e., the C-terminal half of MCW-1) as the antigenic target for these 26 HG sera. Second, Groups II and III HG sera depleted of immunoreactivity with NC16A2.5 continued to react, albeit somewhat more weakly, with MCW-1. Therefore, Groups I, II, and III HG sera (which account for 34 of the 37 HG sera in this study) contain IgG autoantibodies that react with the N-terminal seven amino acid portion of MCW-1, which we now designate MCW-1A. Minor subsets of the HG sera recognized additional sites outside of MCW-1A, but still within the NC16A domain. For example, the eight HG sera in Groups II and III, in addition to reacting with MCW-1A, also reacted with the C-terminal half of MCW-1. This conclusion was drawn from the results of the immunoadsorption experiment described above, i.e., the reduction of immunoreactivity to MCW-1 after preadsorption of the sera against NC16A2.5. The results of other immunoadsorption experiments revealed the presence of two other HG-associated antigenic sites— one within the region of overlap between NC16A2.5 and NC16A3 and one located within NC16A2.5 but not in either MCW-1 or NC16A3 (this last epitope would appear to span the junction of MCW-1 and the BP180 segment in NC16A3). Based on research from our laboratory and others, there is growing evidence that the BP180 antigen is relevant in the pathogenesis of several autoimmune

AUTOANTIBODY EPITOPES IN HERPES GESTATIONIS

subepithelial blistering disorders, which include, in addition to HG, BP (17), cicatricial pemphigoid (19), linear IgA disease (25), and lichen planus pemphigoides (26). These disorders are characterized by subepithelial (mostly subepidermal) inflammatory blisters and the presence of autoantibodies directed against the BP180 antigen. Both BP and HG autoantibodies react with NC16A segments contained in NC16A2 (MCW-1), A2.5 and A3, and, in addition, some BP sera also reacted with NC16A1. Lichen planus pemphigoides autoantibodies react primarily with a site in the C-terminal portion of NC16A4 (25), and a major autoantibodyreactive region associated with cicatricial pemphigoid is located near the C-terminus of the BP180 protein (19, 27). Thus, at the fine specificity level, HG autoantibodies are highly similar to BP autoantibodies, but exhibit differences from those produced by cicatricial pemphigoid and lichen planus pemphigoides patients— differences that might account for differences observed at the clinical level. In conclusion, the extracellular domain of BP180 appears to be the primary target of autoantibodies produced by HG patients. Indeed, the sera of all 37 HG patients in this study contained IgG autoantibodies that reacted with a recombinant protein (sec180e) comprising the extracellular domain of BP180. We further showed that, within the limits of the assays employed, no major HG-associated epitopes were detectable on the BP180 ectodomain outside of those located in the membrane-proximal noncollagenous domain, NC16A. Four HG-associated antigenic sites were localized to a 28 amino acid segment of NC16A (AA 507–534), one of which, a seven amino acid stretch designated MCW1A, is recognized by the vast majority of HG patients’ sera. It is noteworthy that MCW-1A is a subfragment of the 14 amino acid stretch (MCW-1) that was recently shown by our research group to be targeted by both autoantibodies and T cells from an HG patient (22). Taken together, these results provide support for the hypothesis that the NC16A domain of BP180 harbors the key epitope (possibly MCW-1A) that triggers the autoimmune response in HG patients. This molecular information may well aid in the development of more accurate diagnostic assays and more effective therapeutic strategies than those that are currently available for this disease. ACKNOWLEDGMENTS This work was supported in part by U.S. Public Health Service Grants R01-AR32599 and R37-AR32081(L.A.D.) and R01-AR40410 (G.J.G.) from the National Institutes of Health, a VA Merit Review Grant (L.A.D.), and by a Dermatology Foundation Career Development Award (M.S.L.).

291

REFERENCES 1. Shornick, J. K., Bangert, J. L., Freeman, R. G., and Gilliam, J. N., Herpes gestationis: Clinical and histological features of twenty-eight cases. J. Am. Acad. Dermatol. 8, 214 –224, 1983. 2. Provost, T. T., and Tomasi, T. B., Evidence for complement activation via the alternative pathway in skin diseases. I. Herpes gestationis, systemic lupus erythematosus and bullous pemphigoid. J. Clin. Invest. 51, 1779 –1787, 1973. 3. Jordon, R. E., Heine, K. G., Tappeiner, G., Bushkell, L. L., and Provost, T. T., The immunopathology of herpes gestationis: Immunofluorescent studies and characterization of the “HG” factor. J. Clin. Invest. 57, 1426 –1431, 1976. 4. Katz, S. I., Hertz, K. C., and Yaoita, H., Herpes gestationis: Immunopathology and characterization of the HG factor. J. Clin. Invest. 57, 1434 –1441, 1976. 5. Carruthers, J. A., and Ewins, A. R., Herpes gestationis: Studies on the binding characteristics, activity, and pathogenetic significance of the complement-fixing factor. Clin. Exp. Immunol. 31, 38 – 44, 1978. 6. Kelly, S. E., Cerio, R., Bhogal, B. S., and Black, M. M., The distribution of IgG subclasses in pemphigoid gestationis: PG factor is an IgG1 autoantibody. J. Invest. Dermatol. 92, 695– 698, 1989. 7. Labib, R. S., Anhalt, G. J., Patel, H. P., Mutasim, D. F., and Diaz, L. A., Molecular heterogeneity of the bullous pemphigoid antigens as detected by immunoblotting. J. Immunol. 136, 1231– 1235, 1986. 8. Morrison, L. H., Labib, R. S., Zone, J. J., Diaz, L. A., and Anhalt, G. J., Herpes gestationis autoantibodies recognize a 180-kD human epidermal antigen. J. Clin. Invest. 81, 2023–2026, 1988. 9. Diaz, L. A., Ratire, H., III., Sauders, W. S., Futamura, S., Squiquera, H. L., Anhalt, G. J., and Giudice, G. J., Isolation of a human epidermal cDNA corresponding to the 180-kD autoantigen recognized by bullous pemphigoid and herpes gestationis sera. Immunolocalization of this protein to the hemidesmosome. J. Clin. Invest. 86,1088 –1094, 1990. 10. Liu, Z., Diaz, L. A., Troy, J. L., Taylor, A. F., Emery, D. J., Fairley, J. A., and Giudice, G. J., A passive transfer model of the organ-specific autoimmune disease, bullous pemphigoid, using antibodies generated against the hemidesmosomal antigen, BP180. J. Clin. Invest. 92, 2480 –2488, 1993. 11. Liu, Z., Giudice, G. J., Swartz, S. J., Fairley, J. A., Till, G. O., Troy, J. L., and Diaz, L. A., The role of complement in experimental bullous pemphigoid. J. Clin. Invest. 95, 1539 –1544, 1995. 12. Liu, Z., Giudice, G. J., Zhou, X., Swartz, S. J., Troy, J. L., Fairley, J. A., Till, G. O., and Diaz, L. A., A major role for neutrophils in experimental bullous pemphigoid. J. Clin. Invest. 100, 1256 – 1263, 1997. 13. Giudice, G. J., Emery, D. J., Zelickson, B. D., Anhalt, G. J., Liu, Z., and Diaz, L. A., Bullous pemphigoid and herpes gestationis recognize a common non-collagenous site on the BP180 ectodomain. J. Immunol. 151, 5742–5750, 1993. 14. Giudice, G. J., Wilske, K. C., Anhalt, G. J., Fairley, J. A., Taylor, A. F., Emery, D. J., Hoffman, R. G., and Diaz, L. A., Development of an ELISA to detect anti-BP180 autoantibodies in bullous pemphigoid and herpes gestationis. J. Invest. Dermatol. 102, 878 – 881, 1994. 15. Matsumura, K., Amagai, M., Nishikawa, T., and Hashimoto, T., The majority of bullous pemphigoid and herpes gestationis serum samples react with the NC16A domain of the 180-kD bullous pemphigoid antigen. Arch. Dermatol. Res. 288, 507–509, 1996.

292

LIN ET AL.

16. Zillikens, D., Mascaro, J. M., Rose, P. A., Liu, Z., Ewing, S. M., Caux, F., Hoffmann, R. G., Diaz, L. A., and Giudice, G. J., A highly sensitive enzyme-linked immunosorbent assay for the detection of circulating anti-BP180 autoantibodies in patients with bullous pemphigoid. J. Invest. Dermatol. 109, 679 – 683, 1997. 17. Zillikens, D., Rose, P. A., Balding, S. D., Liu, Z., OlagueMarchan, M., Diaz, L. A., and Giudice, G. J., Tight clustering of extracellular BP180 epitopes recognized by bullous pemphigoid autoantibodies. J. Invest. Dermatol. 109, 573–579, 1997. 18. Balding, S. D., Diaz, L. A., and Giudice, G. J., A recombinant form of the human BP180 ectodomain forms a collagen-like homotrimeric complex. Biochemistry 36, 8821– 8830, 1997. 19. Balding, S. D., Prost, C., Diaz, L. A., Bernard, P., Bedane, C., Aberdam, D., and Giudice, G. J., Cicatricial pemphigoid autoantibodies react with multiple sites on the BP180 extracellular domain. J. Invest. Dermatol. 106, 141–146, 1996. 20. Liu, Z., Diaz, L. A., Swartz, S. J., Troy, J. L., Fairley, J. A., and Giudice, G. J., Molecular mapping of pathogenically relevant BP180 epitope associated with experimentally induced murine bullous pemphigoid. J. Immunol. 155, 5449 –5454, 1995. 21. Martins, C. R., Labib, R. S., Rivitti, E. A., and Diaz, L. A., A fragment of pemphigus foliaceus antigen released by trypsinization of viable human epidermis. J. Invest. Dermatol. 95, 208 – 212, 1990. 22. Lin, M. S., Gharia, M., Swartz, S. J., Diaz, L. A., and Giudice, G. J., Identification and characterization of epitopes recognized by T lymphocytes and autoantibodies from patients with herpes gestationis. J. Immunol. 162, 4991– 4997, 1999. Received June 8, 1999; accepted June 16, 1999

23. Yoshikawa, H., Lambert, E. H., Walser-Kuntz, D. R., Yasukawa, Y., McCormick, D. J., and Lennon, V. A., A 17-mer self-peptide of acetylcholine receptor binds to B cell MHC class II, activates helper T cells, and stimulates autoantibody production and electrophysiologic signs of myasthenia gravis. J. Immunol. 159, 1570 –1577, 1997. 24. Wucherpfennig, K. W., Catz, I., Hausmann, S., Strominger, J. L., Steinman, L., and Warren, K. G., Recognition of the immunodominant myelin basic protein peptide by autoantibodies and HLA-DR2-restricted T cell clones from multiple sclerosis patients: Identity of key contact residues in the B-cell and T-cell epitopes. J. Clin. Invest. 100, 1114 –1122, 1997. 25. Herzele, K., Zillikens, D., Schmidt, E., Krenig, H., Mascaro, J. M., Messer, G., Chan, L. S., Olague-Marchan, M., Lin, M. S., Diaz, L. A., Bro¨cker, E. B., and Giudice, G. J., Autoantibodies in a subgroup of patients with linear IgA disease react with the NC16A domain of BP180. J. Invest. Dermatol. 110, 510a, 1998. [Abstract] 26. Zillikens, D., Caux, F., Mascaro, J. M., Prost, C., Krenig, H., Schmidt, E., Callen, J. P., Bro¨cker, E. B., Diaz, L. A., and Giudice, G. J., A novel epitope within the BP180 NC16A domain is targeted by autoantibodies in lichen planus pemphigoides. J. Invest. Dermatol. 110, 514a, 1998. [Abstract] 27. Bedane, C., McMillan, J. R., Balding, S. D., Bernard, P., Prost, C., Bonnetblanc, J. M., Diaz, L. A., Eady, R. A., and Giudice, G. J., Bullous pemphigoid and cicatricial pemphigoid autoantibodies react with ultrastructurally separable epitopes on the BP180 ectodomain: evidence that BP180 spans the lamina lucida. J. Invest. Dermatol. 108, 901–907, 1997.