- Email: [email protected]
Cicatricial pemphigoid with circulating IgA and IgG autoantibodies to the central portion of the BP180 ectodomain: Beneficial effect of adjuvant therapy with high-dose intravenous immunoglobulin
Cicatricial pemphigoid with circulating IgA and IgG autoantibodies to the central portion of the BP180 ectodomain: Beneficial effect of adjuvant therapy with high-dose intravenous immunoglobulin Martin Leverkus,a Matthias Georgi,a Zhuxiang Nie,b Takashi Hashimoto,b Eva-Bettina Bröcker,a and Detlef Zillikensa Würzburg, Germany, and Kurume, Japan Cicatricial pemphigoid (CP) is an autoimmune subepidermal blistering disease characterized by deposits of IgG, IgA, or C3 at the cutaneous basement membrane zone. CP may present with considerable variation regarding age, morphology of lesions, and mucosal involvement, which may heal with or without scarring. We describe a patient with CP who presented with circulating IgA and IgG autoantibodies to the epidermal side of salt-split human skin. By immunoblot analysis, the patient’s IgA reacted with the soluble ectodomain of BP180 (LAD-1). This reactivity was mainly directed to the central portion of the BP180 ectodomain, a site that, to date, has not been described as the target of IgA autoantibodies. Different immunosuppressive treatment regimens including steroids and mycophenolate mofetil did not control this patient’s disease, and severe scarring of the conjunctivae occurred with impairment of vision. Addition of adjuvant intravenous immunoglobulin (1 g/kg body weight on 2 consecutive days) every 4 weeks led to a dramatic improvement of conjunctivitis and gingivitis. Clinical improvement correlated with the serum’s IgA immunoblot reactivity against LAD-1. Further studies on a larger number of patients with CP should try to correlate the specificity of autoantibodies in CP with the response to certain therapeutic regimens. (J Am Acad Dermatol 2002;46:116-22.)
C
icatricial pemphigoid (CP) is a chronic bullous autoimmune disorder that may present with considerable variation regarding age, morphology of lesions, and mucosal involvement that may heal with or without scarring.1 CP is associated with autoantibodies of the IgG or IgA isotype that may be directed to different antigens of the dermoepidermal junction (DEJ).1,2 The best characterized autoantigens of patients with CP are BP180,3 laminin 5,4 type VII collagen,1 and the β4 subunit of α6β4 integrin.5 Other autoantibodies that have been described in patients with CP include uncein6 and a 45-kd protein recognized by IgA autoantibodies in ocular CP.7 Direct immunofluorescence of perilesional mucous
From the Departments of Dermatology, University of Würzburga and University of Kurume.b Reprint requests: Martin Leverkus, Department of Dermatology, University of Würzburg, Josef-Schneider-Str 2, 97080 Würzburg, Germany. Copyright © 2002 by the American Academy of Dermatology, Inc. 0190-9622/2002/$35.00 + 0 16/91/117860 doi:10.1067/mjd.2002.117860
116
membrane or skin biopsy specimens demonstrate linear deposits of IgG or IgA at the DEJ. Indirect immunofluorescence may reveal binding of the autoantibodies to the dermal, epidermal, or both sides of 1 mol/L salt-split human skin. Whereas laminin 5 and type VII collagen appear to represent the major target antigens of dermal binding autoantibodies, BP180 as well as the β4 subunit of α6β4 integrin are recognized by circulating autoantibodies binding to the epidermal side of salt-split skin.2 Circulating IgA autoantibodies may be found in 40% to 60% of sera from patients with the clinical phenotype of CP.8,9 Although the specificity of IgG autoantibodies in CP has been the focus of several studies,3,9,10 target antigens of IgA autoantibodies in patients with CP are less well characterized, possibly because of lower serum titers of these autoantibodies. The treatment of CP may be extremely difficult and ocular involvement may eventually lead to blindness or other severe complications despite aggressive treatment regimens including glucocorticosteroids, cyclosporine, azathioprine, mycophenolate mofetil, and other immunosuppressants.11 High-
Leverkus et al 117
A
➞
➞
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
B Fig 1. A, Inflammation, synechiae, and pannus formation at lateral canthus of left eye (arrow). B, Left eye after 6 cycles of adjuvant therapy with high-dose IVIgs showing less conjunctival injection and regression of pannus formation (arrow).
dose intravenous immunoglobulins (IVIgs) have occasionally been reported to be beneficial in the treatment of patients with subepidermal autoimmune bullous diseases.10,12-15 We describe a patient with the clinical phenotype of CP who demonstrated a major IgA and IgG reactivity to the central portion of the BP180 ectodomain. Treatment with dapsone led to intolerable side effects, and the patient’s disease did not sufficiently respond to high doses of corticosteroids and mycophenolate mofetil. However, after addition of IVIg to the therapeutic regimen, the inflammatory disease of mucous membranes could be stopped.
CASE REPORT A 74-year-old, previously healthy female patient presented with erosions on oral mucous membranes, conjunctivitis, and difficulties swallowing. Her drug history was unremarkable. A biopsy specimen from the oral mucosa showed an erosion with a dense subepithelial infiltrate containing eosinophils and neutrophils. The patient was given prednisolone (1 mg/kg daily) and dapsone (1.5 mg/kg daily). After onset of a maculopapular rash, dapsone was discontinued and mycophenolate mofetil (2 g/d) was added as a corticosteroid-sparing agent. Because of corticosteroid-induced severe dysphoria, prednisolone had to be tapered to 10 mg/d. With this treatment, disease activity was reduced, but lesions did not clear. Six months later, the patient experienced widespread erosions on the gingiva, a marked conjunctivitis, and rapid scarring of the left eye (Fig 1, A). Despite increasing doses of corticosteroids, the disease progressed; treatment with IVIg (1 g/kg body weight daily on 2 consecutive days) was initiated and a total of 6 cycles, each 4 weeks apart, was
administered. Side effects of this treatment were minor and included transient arthralgia and nausea. With the addition of IVIg, conjunctivitis and gingivitis cleared over a period of 6 months (Fig 1, B) and prednisolone could be tapered. Currently, the patient has been without active disease for more than 12 months. Her current medication includes mycophenolate mofetil (2 g/d), prednisone (8 mg/d), and IVIg (1 g/kg body weight on 2 consecutive days every 8 weeks).
MATERIAL AND METHODS Immunofluorescence studies Biopsy specimens from perilesional oral mucosa were studied by direct immunofluorescence microscopy for deposits of IgG, IgA, IgM, and C3. For indirect immunofluorescence microscopy, serum samples of the patient and controls were reacted with 1 mol/L salt-split human skin as described.16 Preparation of epidermal extracts, the soluble BP180 ectodomain (LAD-1) from keratinocyte culture medium, and recombinant BP180 proteins Epidermal extracts were prepared as described.16 For the preparation of the soluble BP180 ectodomain, immortalized human keratinocytes (HaCaT cells, provided by Dr N. Fusenig, German Cancer Research Center [DKFZ], Heidelberg, Germany) were grown in serum-free, low-calcium keratinocyte growth medium (KGM, Cell Systems, St Katharinen, Germany).17 At 70% confluency, Lascorbate (Sigma, Deisenhofen, Germany) was added at a concentration of 100 µg/mL of medium for 48 hours. Proteins of the supernatant were precipitated with 30% ammonium sulfate and resus-
118 Leverkus et al
Fig 2. Schematic diagram of BP180, the keratinocytederived 120-kd protein LAD-1, and recombinant fragments of BP180 used in this study. BP180 is a transmembrane protein of basal keratinocytes with a large ectodomain spanning the lamina lucida of the DEJ. The 120-kd LAD-1 antigen represents a proteolytic cleavage product of the extracellular domain of BP180. The major noncollagenous extracellular domain (NC16A) is located near the cell membrane of basal keratinocytes. Three larger recombinant fusion proteins of the extracellular domain of BP180 are depicted (BP1050, BP963, and BP915). Amino acid numbers are shown at the N- and C-termini of the fragments. The N- and C-termini of LAD-1 are not firmly identified yet. Arrow points to the central portion of the BP180 ectodomain recognized by patient’s IgA autoantibodies.
pended in a buffer containing 25 mmol/L TRIS, pH 7.8, 65 mmol/L NaCl, 5 mmol/L EDTA, and 1 mmol/L aminoethylbenzene-sulfonylfluoride (Sigma).18 After dialyzing against cold resuspension buffer overnight, 5 mmol/L EDTA, 1 mmol/L PMSF, and 1 mmol/L aminoethylbenzene-sulfonylfluoride were added and samples were stored in aliquots at –80°C. Recombinant GST-fusion proteins GST-NC16A, GSTBP1050, GST-BP963, and GST-BP915 were expressed in Escherichia coli.19,20 Proteins were purified by glutathione-agarose affinity chromatography as described.20 Immunoblotting Recombinant and keratinocyte-derived proteins used in this study are shown in Fig 2. Proteins were fractionated by 15% (GST-NC16A), 8% (GSTBP1050/963/915; LAD-1), and 6% (epidermal extract) sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically
J AM ACAD DERMATOL JANUARY 2002
Fig 3. Direct immunofluorescence microscopy of a perilesional skin biopsy specimen. Linear deposits of IgA along DEJ.
Fig 4. Indirect immunofluorescence microscopy on saltsplit human skin. Circulating IgA autoantibodies bind to the epidermal side of split.
transferred to nitrocellulose.20 Blots were blocked for 45 minutes in 3% solution of skimmed milk powder in TBST buffer (0.02 mol/L TRIS-(hydromethyl)aminomethane, 0.14 mol/L NaCl, Tween-20 0.01%, pH 7.5). Human sera were diluted 50-fold in 1% bovine serum albumin (BSA) TBST for detection of IgA and 100-fold for detection of IgG autoantibodies. Rabbit serum R594, directed against the NC16A domain,21 was diluted 1000-fold and rabbit anti-GST antibody (Sigma, Deisenhofen, Germany) 2000-fold in 1% BSA TBST. After a 12-hour incubation with diluted sera at room temperature, the blots were washed and incubated with peroxidase-conjugated
Leverkus et al 119
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
A
B
Fig 5. A, Immunoblot analysis of epidermal extracts. Rabbit serum generated against the NC16A domain of BP180 protein (R594, lane 1) and IgA antibodies of a control patient with LAD (LAD-W, lane 2), but not our patient’s (CP-SC, lane 3) or a normal human serum (NHS, lane 4) recognize a 180-kd protein in epidermal extract (arrow). Markers at the left margin indicate migration positions of molecular weight markers. B, Immunoblot analysis of concentrated medium of HaCaT keratinocytes. Serum from patient CP-SC and a reference control patient with linear IgA disease (LAD-W) demonstrate circulating IgA autoantibodies against the soluble 120-kd ectodomain of BP180 (LAD-1; lanes 2 and 3). Controls include rabbit serum R594 (lane 1) and normal human serum (NHS, lane 4). Markers at the left margin of each panel indicate migration positions of molecular weight markers. Arrow marks the position of LAD-1. To visualize reactivity of R594, goat anti-rabbit IgG (lane 1) was used; for human sera (lanes 2-4), rabbit anti-human IgA was used.
rabbit anti-human IgA and anti-human IgG antibodies (Jackson Immunoresearch Laboratories, West Grove, Pa) or goat anti-rabbit IgG (Dako, Hamburg, Germany) diluted 500-fold in 1% BSA TBST for 1 hour. Diaminobenzidine was used as chromogenic substrate (Merck, Darmstadt, Germany).
RESULTS Immunofluorescence studies Direct immunofluorescence of a perilesional skin biopsy specimen revealed linear deposits of IgA and, to a lesser degree, IgG along the DEJ (Fig 3). Indirect immunofluorescence microscopy demonstrated circulating IgA (1:20) and IgG (1:10) autoantibodies binding to the epidermal side of 1 mol/L salt-split human skin (Fig 4). Characterization of the fine specificity of the patient’s autoantibodies By immunoblot analysis of epidermal extracts, rabbit anti-BP180 serum (R594) and IgA antibodies in
the serum from reference control patients with LAD (LAD-W) or bullous pemphigoid reacted with BP180, whereas our patient’s serum (CP-SC) did not contain IgA (Fig 5, A) or IgG antibodies labeling this protein. In contrast, serum CP-SC contained IgA and, to a lesser extent, IgG autoantibodies reactive with the soluble ectodomain of BP180 (LAD-1) and a 97-kd protein from the conditioned medium of cultured keratinocytes. The same two bands were labeled by IgG from R594 and IgA from control patient LAD-W (Fig 5, B). IgA autoantibodies in the serum of patient LAD-W also reacted with NC16A, whereas serum CPSC was unreactive with this stretch of BP180 (Fig 6). However, IgG autoantibodies from serum CP-SC did react with NC16A (data not shown). To further characterize the fine specificity of our patient’s IgA and IgG autoantibodies to BP180, we performed immunoblot studies using 3 different fragments, together roughly covering the entire BP180 ectodomain (Fig 2). IgA and, to a lesser extent, IgG of CP-SC serum recognized the fusion protein GST-
120 Leverkus et al
J AM ACAD DERMATOL JANUARY 2002
Fig 6. Immunoblot analysis of the recombinant BP180 NC16A domain. IgG antibodies of rabbit serum R594 (lane 1) and IgA autoantibodies of patient LAD-W (lane 2) recognize the recombinant BP180 NC16A domain (arrow), whereas no IgA reactivity to this peptide is found in serum of our patient (CP-SC; lane 3) and normal human serum (NHS, lane 4). Markers at the left margin indicate migration positions of molecular weight markers.
Fig 7. Immunoblot analysis of 3 recombinant fusion proteins together roughly spanning the extracellular domain of BP180. IgA autoantibodies of CP-SC recognize the central portion of the extracellular domain of BP180 (BP963). A rabbit anti-GST antibody reacts with all 3 GST-fusion proteins (arrows). Lower bands of BP1050 represent degradation products detected by anti-GST antibody. Markers at the left margin indicate migration positions of molecular weight markers.
BP963 (encoding the central 321 amino acids [aa] [aa 885-1206]) of the BP 180 ectodomain. In addition, IgG autoantibodies reacted with GST-BP915 (aa 1227-1532), but neither IgA nor IgG reactivity was found with GST-BP1050 (aa 885-1206) (Fig 7).
DISCUSSION
The patient’s IgA immunoblot reactivity to LAD-1 decreased during treatment with IVIg To address the question, if reactivity to BP180 correlates with improvement of the patient’s condition, IgA immunoblot reactivity of the patient’s serum was analyzed during IVIg treatment in monthly intervals. By the time IVIg was initiated, indirect immunofluorescence microscopy and IgG immunoblot reactivity with LAD-1 had been negative. However, there was still a marked IgA reactivity to LAD-1 (Fig 8, lane 1). During the following 6 months, along with clinical improvement, IgA reactivity to LAD-1 decreased (lanes 2-7), while IgG reactivity remained negative.
We describe a patient with the clinical phenotype of CP and circulating IgA and IgG autoantibodies to the BP180 ectodomain. In most patients with CP, the autoimmune response to BP180 has been found to be directed to the C-terminus of the protein.3,19 We characterized the fine specificity of our patient’s autoantibodies using 3 recombinant fragments of BP180 together roughly covering its entire ectodomain. Remarkably, the patient’s IgA autoantibodies recognized only the central portion of the BP180 ectodomain, a site that, to date, has not been described as the target of IgA autoantibodies. IgG autoantibodies of our patient’s serum reacted with the central as well as the C-terminal (BP915) and N-terminal portions (NC16A) of BP180. Studying the autoimmune response in a larger group of patients with CP, Nie and Hashimoto19 recently identified 1 patient with IgG reactivity to the central portion of BP180. However, IgA antibodies of this patient were directed to the C-
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
terminus of BP180. Although our patient’s serum reacted with different recombinant fragments of BP180, no reactivity was detected using keratinocytederived BP180. This discrepancy is likely because the immunoblot assay is more sensitive in detecting antiBP180 antibodies with recombinant BP180 fragments, rather than using BP180 isolated from epidermis or keratinocytes, as previously shown.20,22 In patients with CP, variable degrees of IgA or IgG reactivity have been reported. It appears that a dual IgA and IgG response is associated with a more severe clinical phenotype.8,23,24 Some authors have suggested that this correlation may be explained by the physiologic function of IgA in mucosal immunity, which may lead to increased inflammation in mucosal sites in patients with high titers of pathogenic IgA autoantibodies.25 In contrast to this hypothesis, some patients with CP or linear IgA disease (LAD) demonstrate no mucosal involvement despite high titers of circulating IgA autoantibodies to BP180. Therefore further studies are needed that help to identify individual patients with CP at risk for severe disease or to assign patients with a specific autoantibody reactivity to certain treatment regimens. Application of IVIg has been reported as useful adjuvant therapy for dermatomyositis and several autoimmune bullous diseases including pemphigus vulgaris, bullous pemphigoid, CP, epidermolysis bullosa acquisita, and LAD.12-15,26 The immunomodulatory mechanisms of IVIg may be mediated via the Fc portion of IgG, which interacts with Fc receptors and complement. Other possible mechanisms of action of IVIg include inhibition of complement-mediated damage, neutralization of circulating autoantibodies, the modulation of the production of cytokines and cytokine antagonists, or their inhibition by neutralizing antibodies.27-31 Within 6 cycles of IVIg, inflammatory changes in our patient improved dramatically. We detected a decrease in circulating IgA autoantibodies to the soluble BP180 ectodomain (LAD-1) along with clinical improvement. The mode of action of IVIg in our patient may therefore be attributed, at least in part, to the down-regulation of autoantibody production. Recently, a correlation between clinical response and the titer of IgA autoantibodies to the 97-kd fragment of BP180 was also demonstrated in a patient with LAD treated with adjuvant IVIg.15 In summary, we describe a patient with clinical features of CP and, for the first time, IgA reactivity to the central portion of the extracellular domain of BP180. Adjuvant treatment with IVIg led to a dramatic improvement of the clinical course and correlated well with IgA autoantibody reactivity to BP180. Based on this and previous observations,10,12-15 a multicenter study for the use of IVIg
Leverkus et al 121
Fig 8. Immunoblot reactivity of IgA autoantibodies in our patient’s serum during treatment with high-dose immunoglobulin therapy. Serum samples were reacted with concentrated medium of HaCaT cells containing LAD-1 before (lane 1) and 1, 2, 3, 4, 5, and 6 months after inititation of treatment (lanes 2-7, respectively). Markers at the left margin indicate migration positions of molecular weight markers. Arrow points to position of LAD-1.
in the treatment of recalcitrant autoimmune bullous diseases is warranted. REFERENCES 1. Yancey KB. Cicatricial pemphigoid. In: Freedberg IM, Eisen AZ, Wolff K, Austen KF, Goldsmith LA, Katz SI, et al, editors. Dermatology in general medicine. 5th ed. New York: McGrawHill; 1999. p. 674-9. 2. Zillikens D. Acquired skin disease of hemidesmosomes. J Dermatol Sci 1999;20:134-54. 3. Balding SD, Prost C, Diaz LA, Bernard P, Bedane C, Aberdam D, et al. Cicatricial pemphigoid autoantibodies react with multiple sites on the BP180 extracellular domain. J Invest Dermatol 1996;106:141-6. 4. Domloge-Hultsch N, Anhalt GJ, Gammon WR, Lazarova Z, Briggaman R, Welch M, et al. Anti-epiligrin cicatricial pemphigoid: a subepithelial bullous disorder. Arch Dermatol 1994; 130:1521-9. 5. Tyagi S, Bhol K, Natarajan K, Livir-Rallatos C, Foster CS, Ahmed AR. Ocular cicatricial pemphigoid antigen: partial sequence and biochemical characterization. Proc Natl Acad Sci U S A 1996; 93:14714-9. 6. Horiguchi Y, Ueda M, Shimizu H,Tanaka T, Matsuyoshi N, Utani A, et al. An acquired bullous dermatosis due to an autoimmune reaction against uncein. Br J Dermatol 1996;134:934-8. 7. Smith EP, Taylor TB, Meyer LJ, Zone JJ. Identification of a basement membrane zone antigen reactive with circulating IgA antibody in ocular cicatricial pemphigoid. J Invest Dermatol 1993;101:619-23. 8. Setterfield J, Shirlaw PJ, Kerr-Muir M, Neill S, Bhogal BS, Morgan P, et al. Mucous membrane pemphigoid: a dual circulating antibody response with IgG and IgA signifies a more severe and peristent disease. Br J Dermatol 1998;138:602-10.
122 Leverkus et al
9. Murakami H, Nishioka S, Setterfield J, Bhogal BS, Black MM, Zillikens D, et al. Analysis of antigens targeted by circulating IgG and IgA autoantibodies in 50 patients with cicatricial pemphigoid. J Dermatol Sci 1998;17:39-44. 10. Foster CS, Ahmed AR. Intravenous immunoglobulin therapy for ocular cicatricial pemphigoid:a preliminary study.Ophthalmology 1999;106:2136-43. 11. Axt M,Wever S, Baier G, Bogdan S, Hashimoto T, Brocker EB, et al. Cicatricial pemphigoid: therapeutic problem. Hautarzt 1995; 46:620-7. 12. Meier F, Sonnichsen K, Schaumburg-Lever G, Dopfer R, Rassner G. Epidermolysis bullosa acquisita: efficacy of high-dose intravenous immunoglobulins. J Am Acad Dermatol 1993;29:334-7. 13. Humbert P, Derancourt C, Aubin F, Agache P. Effects of intravenous gamma-globulin in pemphigus. J Am Acad Dermatol 1990;22:326. 14. Godard W, Roujeau JC, Guillot B, Andre C, Rifle G. Bullous pemphigoid and intravenous gammaglobulin. Ann Intern Med 1985;103:964-5. 15. Khan IU, Bhol KC, Ahmed AR. Linear IgA bullous dermatosis in a patient with chronic renal failure: response to intravenous immunoglobulin therapy. J Am Acad Dermatol 1999;40:485-8. 16. Zillikens D, Kawahara Y, Ishiko A, Shimizu H, Mayer J, Rank CV, et al. A novel subepidermal blistering disease with autoantibodies to a 200-kDa antigen of the basement membrane zone. J Invest Dermatol 1996;106:1333-8. 17. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A, Fusenig NE. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol 1988;106:761-71. 18. Marinkovich MP,Taylor TB, Keene DR, Burgeson RE, Zone JJ. LAD1, the linear IgA bullous dermatosis autoantigen, is a novel 120kDa anchoring filament protein synthesized by epidermal cells. J Invest Dermatol 1996;106:734-8. 19. Nie Z, Hashimoto T. IgA antibodies of cicatricial pemphigoid sera specifically react with C-terminus of BP180. J Invest Dermatol 1999;112:254-5. 20. Zillikens D, Rose PA, Balding SD, Liu Z, Olague-Marchan M, Diaz LA, et al. Tight clustering of extracellular BP180 epitopes recognized by bullous pemphigoid autoantibodies. J Invest Dermatol 1997;109:573-9.
J AM ACAD DERMATOL JANUARY 2002
21. Balding SD, Diaz LA, Giudice GJ. A recombinant form of the human BP180 ectodomain forms a collagen-like homotrimeric complex. Biochemistry 1997;36:8821-30. 22. Kromminga A, Scheckenbach C, Georgi M, Hagel C, Arndt R, Christophers E, et al. Patients with bullous pemphigoid and linear IgA disease show a dual IgA and IgG autoimmune response to BP180. J Autoimmun 2000;15:293-300. 23. Chan LS, Hammerberg C, Cooper KD. Cicatricial pemphigoid: identification of two distinct sets of epidermal antigens by IgA and IgG class circulating autoantibodies. Arch Dermatol 1990; 126:1466-8. 24. Hashimoto T, Han-Yaku H, Higashiyama M, Hashimoto K, Iwatsuki K, Kaneko F, et al. Four Japanese patients with cicatricial pemphigoid showing both IgG and IgA antibodies against the 180 kDa bullous pemphigoid antigen. Br J Dermatol 1997; 137:305-6. 25. Egan CA, Taylor TB, Meyer LJ, Petersen MJ, Zone JJ. The immunoglobulin A antibody response in clinical subsets of mucous membrane pemphigoid. Dermatology 1999;198:330-5. 26. Dalakas MC. Intravenous immune globulin therapy for neurologic diseases. Ann Intern Med 1997;126:721-30. 27. Viard I,Wehrli P, Bullani R, Schneider P, Holler N, Salomon D, et al. Inhibition of toxic epidermal necrolysis by blockade of CD95 with human intravenous immunoglobulin. Science 1998;282: 490-3. 28. Dwyer JM. Manipulating the immune system with immune globulin. N Engl J Med 1992;326:107-16. 29. Svenson M, Hansen MB, Bendtzen K. Binding of cytokines to pharmaceutically prepared human immunoglobulin. J Clin Invest 1993;92:2533-9. 30. Andersson J, Skansen-Saphir U, Sparrelid E, Andersson U. Intravenous immune globulin affects cytokine production in T lymphocytes and monocytes/macrophages. Clin Exp Immunol 1996;104(Suppl 1):10-20. 31. Basta M, Dalakas MC. High-dose intravenous immunoglobulin exerts its beneficial effect in patients with dermatomyositis by blocking endomysial deposition of activated complement fragments. J Clin Invest 1994;94:1729-35.
C
icatricial pemphigoid (CP) is a chronic bullous autoimmune disorder that may present with considerable variation regarding age, morphology of lesions, and mucosal involvement that may heal with or without scarring.1 CP is associated with autoantibodies of the IgG or IgA isotype that may be directed to different antigens of the dermoepidermal junction (DEJ).1,2 The best characterized autoantigens of patients with CP are BP180,3 laminin 5,4 type VII collagen,1 and the β4 subunit of α6β4 integrin.5 Other autoantibodies that have been described in patients with CP include uncein6 and a 45-kd protein recognized by IgA autoantibodies in ocular CP.7 Direct immunofluorescence of perilesional mucous
From the Departments of Dermatology, University of Würzburga and University of Kurume.b Reprint requests: Martin Leverkus, Department of Dermatology, University of Würzburg, Josef-Schneider-Str 2, 97080 Würzburg, Germany. Copyright © 2002 by the American Academy of Dermatology, Inc. 0190-9622/2002/$35.00 + 0 16/91/117860 doi:10.1067/mjd.2002.117860
116
membrane or skin biopsy specimens demonstrate linear deposits of IgG or IgA at the DEJ. Indirect immunofluorescence may reveal binding of the autoantibodies to the dermal, epidermal, or both sides of 1 mol/L salt-split human skin. Whereas laminin 5 and type VII collagen appear to represent the major target antigens of dermal binding autoantibodies, BP180 as well as the β4 subunit of α6β4 integrin are recognized by circulating autoantibodies binding to the epidermal side of salt-split skin.2 Circulating IgA autoantibodies may be found in 40% to 60% of sera from patients with the clinical phenotype of CP.8,9 Although the specificity of IgG autoantibodies in CP has been the focus of several studies,3,9,10 target antigens of IgA autoantibodies in patients with CP are less well characterized, possibly because of lower serum titers of these autoantibodies. The treatment of CP may be extremely difficult and ocular involvement may eventually lead to blindness or other severe complications despite aggressive treatment regimens including glucocorticosteroids, cyclosporine, azathioprine, mycophenolate mofetil, and other immunosuppressants.11 High-
Leverkus et al 117
A
➞
➞
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
B Fig 1. A, Inflammation, synechiae, and pannus formation at lateral canthus of left eye (arrow). B, Left eye after 6 cycles of adjuvant therapy with high-dose IVIgs showing less conjunctival injection and regression of pannus formation (arrow).
dose intravenous immunoglobulins (IVIgs) have occasionally been reported to be beneficial in the treatment of patients with subepidermal autoimmune bullous diseases.10,12-15 We describe a patient with the clinical phenotype of CP who demonstrated a major IgA and IgG reactivity to the central portion of the BP180 ectodomain. Treatment with dapsone led to intolerable side effects, and the patient’s disease did not sufficiently respond to high doses of corticosteroids and mycophenolate mofetil. However, after addition of IVIg to the therapeutic regimen, the inflammatory disease of mucous membranes could be stopped.
CASE REPORT A 74-year-old, previously healthy female patient presented with erosions on oral mucous membranes, conjunctivitis, and difficulties swallowing. Her drug history was unremarkable. A biopsy specimen from the oral mucosa showed an erosion with a dense subepithelial infiltrate containing eosinophils and neutrophils. The patient was given prednisolone (1 mg/kg daily) and dapsone (1.5 mg/kg daily). After onset of a maculopapular rash, dapsone was discontinued and mycophenolate mofetil (2 g/d) was added as a corticosteroid-sparing agent. Because of corticosteroid-induced severe dysphoria, prednisolone had to be tapered to 10 mg/d. With this treatment, disease activity was reduced, but lesions did not clear. Six months later, the patient experienced widespread erosions on the gingiva, a marked conjunctivitis, and rapid scarring of the left eye (Fig 1, A). Despite increasing doses of corticosteroids, the disease progressed; treatment with IVIg (1 g/kg body weight daily on 2 consecutive days) was initiated and a total of 6 cycles, each 4 weeks apart, was
administered. Side effects of this treatment were minor and included transient arthralgia and nausea. With the addition of IVIg, conjunctivitis and gingivitis cleared over a period of 6 months (Fig 1, B) and prednisolone could be tapered. Currently, the patient has been without active disease for more than 12 months. Her current medication includes mycophenolate mofetil (2 g/d), prednisone (8 mg/d), and IVIg (1 g/kg body weight on 2 consecutive days every 8 weeks).
MATERIAL AND METHODS Immunofluorescence studies Biopsy specimens from perilesional oral mucosa were studied by direct immunofluorescence microscopy for deposits of IgG, IgA, IgM, and C3. For indirect immunofluorescence microscopy, serum samples of the patient and controls were reacted with 1 mol/L salt-split human skin as described.16 Preparation of epidermal extracts, the soluble BP180 ectodomain (LAD-1) from keratinocyte culture medium, and recombinant BP180 proteins Epidermal extracts were prepared as described.16 For the preparation of the soluble BP180 ectodomain, immortalized human keratinocytes (HaCaT cells, provided by Dr N. Fusenig, German Cancer Research Center [DKFZ], Heidelberg, Germany) were grown in serum-free, low-calcium keratinocyte growth medium (KGM, Cell Systems, St Katharinen, Germany).17 At 70% confluency, Lascorbate (Sigma, Deisenhofen, Germany) was added at a concentration of 100 µg/mL of medium for 48 hours. Proteins of the supernatant were precipitated with 30% ammonium sulfate and resus-
118 Leverkus et al
Fig 2. Schematic diagram of BP180, the keratinocytederived 120-kd protein LAD-1, and recombinant fragments of BP180 used in this study. BP180 is a transmembrane protein of basal keratinocytes with a large ectodomain spanning the lamina lucida of the DEJ. The 120-kd LAD-1 antigen represents a proteolytic cleavage product of the extracellular domain of BP180. The major noncollagenous extracellular domain (NC16A) is located near the cell membrane of basal keratinocytes. Three larger recombinant fusion proteins of the extracellular domain of BP180 are depicted (BP1050, BP963, and BP915). Amino acid numbers are shown at the N- and C-termini of the fragments. The N- and C-termini of LAD-1 are not firmly identified yet. Arrow points to the central portion of the BP180 ectodomain recognized by patient’s IgA autoantibodies.
pended in a buffer containing 25 mmol/L TRIS, pH 7.8, 65 mmol/L NaCl, 5 mmol/L EDTA, and 1 mmol/L aminoethylbenzene-sulfonylfluoride (Sigma).18 After dialyzing against cold resuspension buffer overnight, 5 mmol/L EDTA, 1 mmol/L PMSF, and 1 mmol/L aminoethylbenzene-sulfonylfluoride were added and samples were stored in aliquots at –80°C. Recombinant GST-fusion proteins GST-NC16A, GSTBP1050, GST-BP963, and GST-BP915 were expressed in Escherichia coli.19,20 Proteins were purified by glutathione-agarose affinity chromatography as described.20 Immunoblotting Recombinant and keratinocyte-derived proteins used in this study are shown in Fig 2. Proteins were fractionated by 15% (GST-NC16A), 8% (GSTBP1050/963/915; LAD-1), and 6% (epidermal extract) sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically
J AM ACAD DERMATOL JANUARY 2002
Fig 3. Direct immunofluorescence microscopy of a perilesional skin biopsy specimen. Linear deposits of IgA along DEJ.
Fig 4. Indirect immunofluorescence microscopy on saltsplit human skin. Circulating IgA autoantibodies bind to the epidermal side of split.
transferred to nitrocellulose.20 Blots were blocked for 45 minutes in 3% solution of skimmed milk powder in TBST buffer (0.02 mol/L TRIS-(hydromethyl)aminomethane, 0.14 mol/L NaCl, Tween-20 0.01%, pH 7.5). Human sera were diluted 50-fold in 1% bovine serum albumin (BSA) TBST for detection of IgA and 100-fold for detection of IgG autoantibodies. Rabbit serum R594, directed against the NC16A domain,21 was diluted 1000-fold and rabbit anti-GST antibody (Sigma, Deisenhofen, Germany) 2000-fold in 1% BSA TBST. After a 12-hour incubation with diluted sera at room temperature, the blots were washed and incubated with peroxidase-conjugated
Leverkus et al 119
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
A
B
Fig 5. A, Immunoblot analysis of epidermal extracts. Rabbit serum generated against the NC16A domain of BP180 protein (R594, lane 1) and IgA antibodies of a control patient with LAD (LAD-W, lane 2), but not our patient’s (CP-SC, lane 3) or a normal human serum (NHS, lane 4) recognize a 180-kd protein in epidermal extract (arrow). Markers at the left margin indicate migration positions of molecular weight markers. B, Immunoblot analysis of concentrated medium of HaCaT keratinocytes. Serum from patient CP-SC and a reference control patient with linear IgA disease (LAD-W) demonstrate circulating IgA autoantibodies against the soluble 120-kd ectodomain of BP180 (LAD-1; lanes 2 and 3). Controls include rabbit serum R594 (lane 1) and normal human serum (NHS, lane 4). Markers at the left margin of each panel indicate migration positions of molecular weight markers. Arrow marks the position of LAD-1. To visualize reactivity of R594, goat anti-rabbit IgG (lane 1) was used; for human sera (lanes 2-4), rabbit anti-human IgA was used.
rabbit anti-human IgA and anti-human IgG antibodies (Jackson Immunoresearch Laboratories, West Grove, Pa) or goat anti-rabbit IgG (Dako, Hamburg, Germany) diluted 500-fold in 1% BSA TBST for 1 hour. Diaminobenzidine was used as chromogenic substrate (Merck, Darmstadt, Germany).
RESULTS Immunofluorescence studies Direct immunofluorescence of a perilesional skin biopsy specimen revealed linear deposits of IgA and, to a lesser degree, IgG along the DEJ (Fig 3). Indirect immunofluorescence microscopy demonstrated circulating IgA (1:20) and IgG (1:10) autoantibodies binding to the epidermal side of 1 mol/L salt-split human skin (Fig 4). Characterization of the fine specificity of the patient’s autoantibodies By immunoblot analysis of epidermal extracts, rabbit anti-BP180 serum (R594) and IgA antibodies in
the serum from reference control patients with LAD (LAD-W) or bullous pemphigoid reacted with BP180, whereas our patient’s serum (CP-SC) did not contain IgA (Fig 5, A) or IgG antibodies labeling this protein. In contrast, serum CP-SC contained IgA and, to a lesser extent, IgG autoantibodies reactive with the soluble ectodomain of BP180 (LAD-1) and a 97-kd protein from the conditioned medium of cultured keratinocytes. The same two bands were labeled by IgG from R594 and IgA from control patient LAD-W (Fig 5, B). IgA autoantibodies in the serum of patient LAD-W also reacted with NC16A, whereas serum CPSC was unreactive with this stretch of BP180 (Fig 6). However, IgG autoantibodies from serum CP-SC did react with NC16A (data not shown). To further characterize the fine specificity of our patient’s IgA and IgG autoantibodies to BP180, we performed immunoblot studies using 3 different fragments, together roughly covering the entire BP180 ectodomain (Fig 2). IgA and, to a lesser extent, IgG of CP-SC serum recognized the fusion protein GST-
120 Leverkus et al
J AM ACAD DERMATOL JANUARY 2002
Fig 6. Immunoblot analysis of the recombinant BP180 NC16A domain. IgG antibodies of rabbit serum R594 (lane 1) and IgA autoantibodies of patient LAD-W (lane 2) recognize the recombinant BP180 NC16A domain (arrow), whereas no IgA reactivity to this peptide is found in serum of our patient (CP-SC; lane 3) and normal human serum (NHS, lane 4). Markers at the left margin indicate migration positions of molecular weight markers.
Fig 7. Immunoblot analysis of 3 recombinant fusion proteins together roughly spanning the extracellular domain of BP180. IgA autoantibodies of CP-SC recognize the central portion of the extracellular domain of BP180 (BP963). A rabbit anti-GST antibody reacts with all 3 GST-fusion proteins (arrows). Lower bands of BP1050 represent degradation products detected by anti-GST antibody. Markers at the left margin indicate migration positions of molecular weight markers.
BP963 (encoding the central 321 amino acids [aa] [aa 885-1206]) of the BP 180 ectodomain. In addition, IgG autoantibodies reacted with GST-BP915 (aa 1227-1532), but neither IgA nor IgG reactivity was found with GST-BP1050 (aa 885-1206) (Fig 7).
DISCUSSION
The patient’s IgA immunoblot reactivity to LAD-1 decreased during treatment with IVIg To address the question, if reactivity to BP180 correlates with improvement of the patient’s condition, IgA immunoblot reactivity of the patient’s serum was analyzed during IVIg treatment in monthly intervals. By the time IVIg was initiated, indirect immunofluorescence microscopy and IgG immunoblot reactivity with LAD-1 had been negative. However, there was still a marked IgA reactivity to LAD-1 (Fig 8, lane 1). During the following 6 months, along with clinical improvement, IgA reactivity to LAD-1 decreased (lanes 2-7), while IgG reactivity remained negative.
We describe a patient with the clinical phenotype of CP and circulating IgA and IgG autoantibodies to the BP180 ectodomain. In most patients with CP, the autoimmune response to BP180 has been found to be directed to the C-terminus of the protein.3,19 We characterized the fine specificity of our patient’s autoantibodies using 3 recombinant fragments of BP180 together roughly covering its entire ectodomain. Remarkably, the patient’s IgA autoantibodies recognized only the central portion of the BP180 ectodomain, a site that, to date, has not been described as the target of IgA autoantibodies. IgG autoantibodies of our patient’s serum reacted with the central as well as the C-terminal (BP915) and N-terminal portions (NC16A) of BP180. Studying the autoimmune response in a larger group of patients with CP, Nie and Hashimoto19 recently identified 1 patient with IgG reactivity to the central portion of BP180. However, IgA antibodies of this patient were directed to the C-
J AM ACAD DERMATOL VOLUME 46, NUMBER 1
terminus of BP180. Although our patient’s serum reacted with different recombinant fragments of BP180, no reactivity was detected using keratinocytederived BP180. This discrepancy is likely because the immunoblot assay is more sensitive in detecting antiBP180 antibodies with recombinant BP180 fragments, rather than using BP180 isolated from epidermis or keratinocytes, as previously shown.20,22 In patients with CP, variable degrees of IgA or IgG reactivity have been reported. It appears that a dual IgA and IgG response is associated with a more severe clinical phenotype.8,23,24 Some authors have suggested that this correlation may be explained by the physiologic function of IgA in mucosal immunity, which may lead to increased inflammation in mucosal sites in patients with high titers of pathogenic IgA autoantibodies.25 In contrast to this hypothesis, some patients with CP or linear IgA disease (LAD) demonstrate no mucosal involvement despite high titers of circulating IgA autoantibodies to BP180. Therefore further studies are needed that help to identify individual patients with CP at risk for severe disease or to assign patients with a specific autoantibody reactivity to certain treatment regimens. Application of IVIg has been reported as useful adjuvant therapy for dermatomyositis and several autoimmune bullous diseases including pemphigus vulgaris, bullous pemphigoid, CP, epidermolysis bullosa acquisita, and LAD.12-15,26 The immunomodulatory mechanisms of IVIg may be mediated via the Fc portion of IgG, which interacts with Fc receptors and complement. Other possible mechanisms of action of IVIg include inhibition of complement-mediated damage, neutralization of circulating autoantibodies, the modulation of the production of cytokines and cytokine antagonists, or their inhibition by neutralizing antibodies.27-31 Within 6 cycles of IVIg, inflammatory changes in our patient improved dramatically. We detected a decrease in circulating IgA autoantibodies to the soluble BP180 ectodomain (LAD-1) along with clinical improvement. The mode of action of IVIg in our patient may therefore be attributed, at least in part, to the down-regulation of autoantibody production. Recently, a correlation between clinical response and the titer of IgA autoantibodies to the 97-kd fragment of BP180 was also demonstrated in a patient with LAD treated with adjuvant IVIg.15 In summary, we describe a patient with clinical features of CP and, for the first time, IgA reactivity to the central portion of the extracellular domain of BP180. Adjuvant treatment with IVIg led to a dramatic improvement of the clinical course and correlated well with IgA autoantibody reactivity to BP180. Based on this and previous observations,10,12-15 a multicenter study for the use of IVIg
Leverkus et al 121
Fig 8. Immunoblot reactivity of IgA autoantibodies in our patient’s serum during treatment with high-dose immunoglobulin therapy. Serum samples were reacted with concentrated medium of HaCaT cells containing LAD-1 before (lane 1) and 1, 2, 3, 4, 5, and 6 months after inititation of treatment (lanes 2-7, respectively). Markers at the left margin indicate migration positions of molecular weight markers. Arrow points to position of LAD-1.
in the treatment of recalcitrant autoimmune bullous diseases is warranted. REFERENCES 1. Yancey KB. Cicatricial pemphigoid. In: Freedberg IM, Eisen AZ, Wolff K, Austen KF, Goldsmith LA, Katz SI, et al, editors. Dermatology in general medicine. 5th ed. New York: McGrawHill; 1999. p. 674-9. 2. Zillikens D. Acquired skin disease of hemidesmosomes. J Dermatol Sci 1999;20:134-54. 3. Balding SD, Prost C, Diaz LA, Bernard P, Bedane C, Aberdam D, et al. Cicatricial pemphigoid autoantibodies react with multiple sites on the BP180 extracellular domain. J Invest Dermatol 1996;106:141-6. 4. Domloge-Hultsch N, Anhalt GJ, Gammon WR, Lazarova Z, Briggaman R, Welch M, et al. Anti-epiligrin cicatricial pemphigoid: a subepithelial bullous disorder. Arch Dermatol 1994; 130:1521-9. 5. Tyagi S, Bhol K, Natarajan K, Livir-Rallatos C, Foster CS, Ahmed AR. Ocular cicatricial pemphigoid antigen: partial sequence and biochemical characterization. Proc Natl Acad Sci U S A 1996; 93:14714-9. 6. Horiguchi Y, Ueda M, Shimizu H,Tanaka T, Matsuyoshi N, Utani A, et al. An acquired bullous dermatosis due to an autoimmune reaction against uncein. Br J Dermatol 1996;134:934-8. 7. Smith EP, Taylor TB, Meyer LJ, Zone JJ. Identification of a basement membrane zone antigen reactive with circulating IgA antibody in ocular cicatricial pemphigoid. J Invest Dermatol 1993;101:619-23. 8. Setterfield J, Shirlaw PJ, Kerr-Muir M, Neill S, Bhogal BS, Morgan P, et al. Mucous membrane pemphigoid: a dual circulating antibody response with IgG and IgA signifies a more severe and peristent disease. Br J Dermatol 1998;138:602-10.
122 Leverkus et al
9. Murakami H, Nishioka S, Setterfield J, Bhogal BS, Black MM, Zillikens D, et al. Analysis of antigens targeted by circulating IgG and IgA autoantibodies in 50 patients with cicatricial pemphigoid. J Dermatol Sci 1998;17:39-44. 10. Foster CS, Ahmed AR. Intravenous immunoglobulin therapy for ocular cicatricial pemphigoid:a preliminary study.Ophthalmology 1999;106:2136-43. 11. Axt M,Wever S, Baier G, Bogdan S, Hashimoto T, Brocker EB, et al. Cicatricial pemphigoid: therapeutic problem. Hautarzt 1995; 46:620-7. 12. Meier F, Sonnichsen K, Schaumburg-Lever G, Dopfer R, Rassner G. Epidermolysis bullosa acquisita: efficacy of high-dose intravenous immunoglobulins. J Am Acad Dermatol 1993;29:334-7. 13. Humbert P, Derancourt C, Aubin F, Agache P. Effects of intravenous gamma-globulin in pemphigus. J Am Acad Dermatol 1990;22:326. 14. Godard W, Roujeau JC, Guillot B, Andre C, Rifle G. Bullous pemphigoid and intravenous gammaglobulin. Ann Intern Med 1985;103:964-5. 15. Khan IU, Bhol KC, Ahmed AR. Linear IgA bullous dermatosis in a patient with chronic renal failure: response to intravenous immunoglobulin therapy. J Am Acad Dermatol 1999;40:485-8. 16. Zillikens D, Kawahara Y, Ishiko A, Shimizu H, Mayer J, Rank CV, et al. A novel subepidermal blistering disease with autoantibodies to a 200-kDa antigen of the basement membrane zone. J Invest Dermatol 1996;106:1333-8. 17. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A, Fusenig NE. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol 1988;106:761-71. 18. Marinkovich MP,Taylor TB, Keene DR, Burgeson RE, Zone JJ. LAD1, the linear IgA bullous dermatosis autoantigen, is a novel 120kDa anchoring filament protein synthesized by epidermal cells. J Invest Dermatol 1996;106:734-8. 19. Nie Z, Hashimoto T. IgA antibodies of cicatricial pemphigoid sera specifically react with C-terminus of BP180. J Invest Dermatol 1999;112:254-5. 20. Zillikens D, Rose PA, Balding SD, Liu Z, Olague-Marchan M, Diaz LA, et al. Tight clustering of extracellular BP180 epitopes recognized by bullous pemphigoid autoantibodies. J Invest Dermatol 1997;109:573-9.
J AM ACAD DERMATOL JANUARY 2002
21. Balding SD, Diaz LA, Giudice GJ. A recombinant form of the human BP180 ectodomain forms a collagen-like homotrimeric complex. Biochemistry 1997;36:8821-30. 22. Kromminga A, Scheckenbach C, Georgi M, Hagel C, Arndt R, Christophers E, et al. Patients with bullous pemphigoid and linear IgA disease show a dual IgA and IgG autoimmune response to BP180. J Autoimmun 2000;15:293-300. 23. Chan LS, Hammerberg C, Cooper KD. Cicatricial pemphigoid: identification of two distinct sets of epidermal antigens by IgA and IgG class circulating autoantibodies. Arch Dermatol 1990; 126:1466-8. 24. Hashimoto T, Han-Yaku H, Higashiyama M, Hashimoto K, Iwatsuki K, Kaneko F, et al. Four Japanese patients with cicatricial pemphigoid showing both IgG and IgA antibodies against the 180 kDa bullous pemphigoid antigen. Br J Dermatol 1997; 137:305-6. 25. Egan CA, Taylor TB, Meyer LJ, Petersen MJ, Zone JJ. The immunoglobulin A antibody response in clinical subsets of mucous membrane pemphigoid. Dermatology 1999;198:330-5. 26. Dalakas MC. Intravenous immune globulin therapy for neurologic diseases. Ann Intern Med 1997;126:721-30. 27. Viard I,Wehrli P, Bullani R, Schneider P, Holler N, Salomon D, et al. Inhibition of toxic epidermal necrolysis by blockade of CD95 with human intravenous immunoglobulin. Science 1998;282: 490-3. 28. Dwyer JM. Manipulating the immune system with immune globulin. N Engl J Med 1992;326:107-16. 29. Svenson M, Hansen MB, Bendtzen K. Binding of cytokines to pharmaceutically prepared human immunoglobulin. J Clin Invest 1993;92:2533-9. 30. Andersson J, Skansen-Saphir U, Sparrelid E, Andersson U. Intravenous immune globulin affects cytokine production in T lymphocytes and monocytes/macrophages. Clin Exp Immunol 1996;104(Suppl 1):10-20. 31. Basta M, Dalakas MC. High-dose intravenous immunoglobulin exerts its beneficial effect in patients with dermatomyositis by blocking endomysial deposition of activated complement fragments. J Clin Invest 1994;94:1729-35.