Complement fixation by pemphigus antibody. V. Assembly of the membrane attack complex on cultured human keratinocytes.

Previous studies have shown that pemphigus vulgaris (PV) IgG will fix early complement components (C1q, C4, and C3) to cultured murine epidermal cell surfaces and that PV IgG and complement alter epidermal cell membrane integrity. The present study was undertaken to determine if assembly of terminal complement components (C5, C6, C7, C8, and C9) and expression of C5b-9 neoantigens occur when PV IgG interacts with human keratinocyte (HuK) cell surface antigens in the presence of a source of complement. Monoclonal antibodies specific for C5, C6, C7, C8, C9, and C5b-9 neoantigens were screened for reactivity to the individual complement components in an assembled complex of human C5b-9 on rabbit red blood cell ghosts. Monoclonal antibodies (tissue culture supernatants) that bound to antigenic determinants accessible in the C5b-9 complex were selected for this study using immunofluorescence methods. HuK treated with PV IgG fixed C5, C6, C7, C8, C9, and C5b-9 neoantigens in a characteristic speckled pattern, while normal IgG did not. Heat inactivation or EDTA treatment of the complement source, or substitution of C2-depleted serum abolished C5, C6, C7, C8, C9, and C5b-9 neoantigen staining. PV IgG and complement also resulted in significant cytotoxicity to cell membranes as assessed using an ethidium bromide-fluorescein diacetate assay. These results suggest that PV IgG will activate the membrane attack complex of the complement system on HuK cell surfaces, resulting in cytotoxicity to cell membranes, further implicating complement in the pathogenesis of pemphigus.

The complement system in bullous pemphigoid. I. Complement and component levels in sera and blister fluids.

Compared with other serum and blister fluid proteins, total hemolytic complement was reduced in the blister fluid of six serologically positive bullous pemphigold patients while four serologically negative cases had blister fluid complement levels closely approaching the serum levels. Except for pemphigus vulgaris blisters. blister fluids from most patients with other bullous diseases and experimentally induced blisters had blister fluid complement levels more closely approaching the serum levels. With the exception of the two terminal components. C8 and C9, individual components of the complement sequence were also reduced in the blister fluids of the six bullous pemphigold patients with circulating basement membrane zone antibodies. On the other hand, transferrin and IgG levels of these same six serologically positive blister fluids closely approached the corresponding serum levels. Conversion of C3 proactivator was also demonstrable in the serologically positive bullous pemphigoid blister fluids, but not in the corresponding sera. Our studies, therefore, are suggestive of local activation of the complement sequence, by both the classical and alternate pathways, in blisters of serologically positive bullous pemphigold patients.

The late phase of the immediate wheal and flare skin reaction. Its dependence upon IgE antibodies.

IgE antibodies are usually thought to induce only immediate skin reactions. We have shown that the intradermal injection of a number of different allergens can produce a prolonged inflammatory reaction after the immediate wheal and flare in most sensitive subjects. This late inflammatory response occurs 6-12 h after challenge and is characterized by diffuse edema, erythema, pruritus, and heat. Both immediate and late responses can also be seen after passive sensitization of skin sites in nonatopic subjects. That IgE is involved in inducing the reaction was shown by the abolition of both immediate and late responses by passive transfer tests in the following experiments: (a) heating atopic serum at 56degreesC for 4 h, (b) removing IgE from the atopic serum by a solid phase anti-IgE immunoabsorbent, and (c) competitively inhibiting the binding of IgE antibodies to cells by an IgE myeloma protein. In addition, both responses were induced by affinity chromatography-purified IgE antibody, followed by antigenic challenge. Very similar lesions could also be induced by intradermal injection of Compound 48/80, thus suggesting a central role in the reaction for the mast cell or basophil. Histologically, the late phase is characterized by edema and a mixed cellular infiltration, predominantly lymphocytic but also containing eosinophils, neutrophils and basophils. Direct immunofluorescent staining did not show deposition of immunoglobulins or complement components, except IgM in 2 of 15 and C3 in 1 of 15 patients. This finding indicates that the late phase does not depend on the deposition of immune complexes. The results of the study suggest that IgE-allergen interaction on the surfaces of mast cells or on infiltrating basophils causes both immediate and late cutaneous responses.

The immunopathology of herpes gestationis. Immunofluorescence studies and characterization of "HG factor".

Nine skin biopsies from seven herpes gestationis patients were studied by immunofluorescence (IF) techniques. Basement membrane zone (BMZ) deposition of C3 and properdin was present in all nine skin specimens, while IgG deposition was apparent in only one. With in vitro C3 IF staining, positive BMZ staining (HG factor activity) was noted with all seven of our patients' serum samples tested. By standard indirect IF staining, however, only one of these serum samples contained BMZ antibodies of the IgG type. Two cord serum samples, tested by these same methods, yielded positive in vitro C3 staining (HG factor activity) but negative indirect IF staining (IgG). HG factor activity was found to be stable at 56 degrees C for 30 min and in two of three specimens at 56 degrees C for 1 h. Treatment of the complement source (normal human serum) used in the in vitro C3 staining assay with Mg2-EGTA or use of C2-deficient serum as the complement source inhibited HG factor activity. HG factor blocked the specific staining of the BMZ of normal human skin by labeled bullous pemphigoid antibodies. By sucrose density gradient ultracentrifugation and gel chromatography (Sephadex G-200), HG factor activity eluted with IgG-containing fractions. The highly purified IgG fraction of two herpes gestationis sera was also positive for HG factor activity. Our studies suggest that HG factor is an IgG antibody that may not be demonstrable by conventional IF methods, but which activates the classical complement pathway.

Complement activation in bullous skin diseases.

Pemphigus, bullous pemphigoid, cicatricial pemphigoid, dermatitis herpetiformis, and herpes gestationis are members of the chronic vesiculobullous skin diseases of man. The complement system, including both the classical and alternative pathways, may be important in the pathogenesis of these diseases. In pemphigus, early complement components (C1, C4, and C2) appear to be activated in addition to later components (C3 and C5), suggestive of classical pathway activation. Participation of properdin in addition to early complement components suggests local activation of both complement pathways in bullous pemphigoid and cicatricial pemphigoid. Herpes gestationis and dermatitis herpetiformis may be bullous skin diseases entirely mediated by the alternate or properdin pathway. The specific immunopathologic findings in these diseases are discussed.

Immunohistochemical studies of the human cutaneous basement membrane-anchoring fibril complex.

Antiserum to the human cutaneous basement membrane-anchoring fibril complex specifically binds to the basement membrane zone of normal human skin as demonstrated by indirect immunofluorescence and, in addition, cross-reacts with normal human glomerulus. This antiserum does not prevent binding of pemphigoid antibody to its specific antigen; thus the antigen of bullous pemphigoid is unlikely to be a component of the cutaneous basement membrane (basal lamina).

The complement system in bullous pemphigoid: VII. Fixation of the regulatory protein beta 1H globulin by pemphigoid antibody.

Using in vitro complement immunofluorescent staining methods, serum samples from 5 active cases of bullous pemphigoid, with pemphigoid antibody titers of 320 or greater, were tested for their ability to fix the regulatory protein beta 1H globulin in addition to C4 and C3. All 5 samples yielded positive C3, C4 and beta 1H staining reactions in a linear fashion along the basement membrane zone. Heat inactivation or treatment of the complement source (fresh normal human serum) with EDTA, Mg2-EGTA abolished all 3 staining reactions. Substitution of C2-deficient serum as the source of complement inhibited both C3 and beta 1H staining but had no effect on C4 staining. Use of serum devoid of beta 1H (R beta 1H) minimally enhanced C3 staining while no beta 1H staining was observed. The addition of beta 1H to R beta 1H restored positive beta 1H staining. Skin biopsies of perilesional skin from 6 patients with bullous pemphigoid demonstrated heavy in vivo deposition of beta 1H in addition to C3. These studies suggest that pemphigoid antibodies will fix the regulatory protein beta 1H in addition to other complement components, a phenomenon which requires activation of the classical complement pathway and generation of the C3b amplification convertase.

Immunopathologic mechanisms in pemphigus and bullous pemphigoid.

Pemphigus and bullous pemphigoid are autoimmune bullous diseases of the skin. Pemphigus, an intraepidermal blistering disease, is characterized by autoantibodies reactive with antigens located in the intercellular spaces or on the surfaces of epidermal cells. These antibodies, which have recently been shown to activate complement, appear to be the cause of the basic pathologic process of pemphigus, acantholysis. The complement system and the plasminogen-plasmin system may be important mediators in the detachment of epidermal cells. Bullous pemphigoid, a subepidermal blistering disease, is characterized by autoantibodies reactive with an antigen located in the lamina lucida region of the basement membrane zone. These autoantibodies, which will avidly fix complement, appear to mediate subepidermal separation by attraction of a variety of inflammatory cells. Anaphylatoxins, released by activation of C4 and C3, or specific IgE antibodies, may activate mast cells with release of ECF-A attracting eosinophils. With activation of C5, C5a is released which could attract polymorphonuclear leukocytes. Antigen-specific lymphocytes, which can also contribute histamine releasing substances, may also be involved. The exact mechanism by which the epidermis separates from the dermis in bullous pemphigoid, however, remains unresolved.

Complement fixation by pemphigus antibody. III. Altered epidermal cell membrane integrity mediated by pemphigus antibody and complement.

The present study investigates the effects of pemphigus IgG and complement upon cell viability and/or membrane integrity using trypan blue exclusion, ethidium bromide (EB) staining, and fluorescein diacetate (FDA) conversion by living cells. Forty-eight-hour cultivated epidermal monolayers of neonatal BALB/c mice were incubated in media containing 1 mg/ml purified pemphigus IgG for 48 h in either the presence or absence of complement (absorbed AB sera). Adherent and detached cells were examined by both phase and fluorescence microscopy. Results from trypan blue exclusion showed that pemphigus IgG plus complement produced a modest decrease in exclusion of the dye compared to pemphigus IgG without complement. When FDA/EB comparisons were made, however, the differences were more substantial. When complement plus pemphigus IgG was added to cultures, the number of FDA-positive adherent cells decreased significantly and the number of EB-positive detached cells increased significantly. The effects of complement were inhibited by the use of heat-inactivated AB sera or by C1q depletion of AB sera. No significant effect on the cells was observed in the presence or absence of complement when pemphigus F(ab')2 fragments or when normal IgG was used. Plasminogen depletion of the complement source did not interfere with complement and pemphigus IgG effects as judged by the FDA/EB assay. These studies suggest that pemphigus antibody in the presence of complement alters cell membrane integrity and supports the contention that complement may play a significant role in the mechanism of acantholysis.

Clq binding activity in lupus erythematosus: correlation with the "lupus band test".

A 131I C1q binding assay was employed to estimate C1q binding activity (C1q BA), presumably due to the presence of circulating immune complexes, in lupus erythematosus sera. Thirteen of 16 sera from patients with systemic lupus erythematosus (SLE) had elevated C1q BA. Only 1 of 17 sera from patients with generalized discoid LE (GDLE) and none of 5 localized DLE patient's sera had elevated C1q BA. A correlation between a positive LE band test and the level of C1q BA was apparent. A positive LE band test tended to occur in patients with higher levels of C1q BA, whereas a negative LE band test tended to occur in patients with low levels of C1q BA.

Complement fixation by pemphigus antibody. I. In vitro fixation to organ and tissue culture skin.

Although complement is often detected in the intercellular substance of pemphigus skin lesions, the ability of pemphigus antibodies to fix complement in vitro is controversial. The purpose of this study was to test in vitro complement fixation abilities of pemphigus antibodies further using organ and tissue culture methods. Epidermal cell monolayers from mouse tail were incubated with the purified IgG fraction of pemphigus serum followed by purified Clq. Binding of Clq, as well as IgG was demonstrated by immunofluorescence methods. When purified Clq was replaced with normal human serum as a complement source, positive C3 and C4 staining were also evident. When purified IgG of normal human serum was used in place of pemphigus IgG, similar immunofluorescence staining was not observed. Further evidence for complement fixation in vitro by pemphigus antibodies was obtained using organ cultures. Organ culture of normal human skin and monkey esophageal mucosa cultured in purified pemphigus IgG showed intercellular substance binding of IgG. No binding was observed when normal IgG was substituted for pemphigus IgG. Additional organ culture sections were then treated with complement (fresh normal human serum) and tested by in vitro complement staining. Fixation of Clq, C4, and C3 was noted in intercellular substance areas of organ cultured skin and mucosa incubated with pemphigus IgG but not those incubated with normal IgG. Prior treatment of pemphigus IgG organ cultured skin sections with unlabeled anti-C3, blocked positive C3 staining. These results suggest that some pemphigus antibodies are capable of activating complement in vitro.

Deposition of the membrane attack complex of complement in pemphigus vulgaris and pemphigus foliaceus skin.

The present study was performed to determine whether complement activation in pemphigus vulgaris (PV) and pemphigus foliaceus (PF) results in the assembly of the terminal complement sequence or membrane attack complex (MAC) in skin lesions. Biopsy specimens of skin lesions from five patients with PV and three patients with PF contained C5, C7, C9, and the MAC related neoantigen (C5b-9 neoantigen) in intercellular substance areas (ICS), as well as IgG and the early complement components Clq, C4, and C3. The presence of these late complement components and the C5b-9 neoantigens in ICS sites of the skin lesions is indicative of complement activation by the pemphigus antibody, with subsequent assembly of the MAC. The binding of IgG and early complement components to ICS was observed in both non-lesional (normal appearing) skin and in skin lesions. However, no MAC could be detected in the normal appearing skin of our pemphigus patients. It was also noted that the MAC could be generated in vitro on cryostat sectioned normal human skin by pemphigus antibody in the presence of complement. Results of these studies suggest that complement activation may be related to membrane damage of epidermal cells in both PV and PF.

Erythema multiforme: direct immunofluorescence studies and detection of circulating immune complexes.

The immunologic parameters of 23 patients with erythema multiforme who were seen by us (17 patients) or who had biopsies sent for immunofluorescence testing (6 cases) are reviewed. Biopsy specimens were sectioned and tested with labeled antisera to human IgG, IgA, IgM, C3 and fibrin. Fourteen biopsies showed IgM deposits in the superficial blood vessels, 13 demonstrated C3, 15 showed fibrin deposition, and 1 biopsy showed IgA deposition. All biopsies were negative for IgG. Eight serum samples tested by indirect IF were negative for skin-reactive antibodies. In addition to IF testing, serum samples from 20 patients were tested for circulating immune complexes with a Clq binding radioassay and a monoclonal rheumatoid factor (mRF) inhibition assay. Immune complexes were not detected by the Clq binding assay, but 6 of 20 serum samples demonstrated low to moderate levels of immune complexes by the mRF inhibition assay. By sucrose density gradient ultracentrifugation in the mRF-reactive material in one serum sample sedimented in high molecular weight fractions and also demonstrated anticomplementary activity. These findings suggest that immune complex formation and subsequent deposition in the cutaneous microvasculature may play a role in the pathogenesis of erythema multiforme.

Inflammation in acne vulgaris: leukocyte attraction and cytotoxicity by comedonal material.

The chemoattraction of comedonal material for leukocytes was evaluated. Material from open comedones attracted mononculear leukocytes but did not attract polymorphonuclear leukocytes. At higher concentrations, comedonal material was cytotoxic for leukocytes of both types. Of the comedonal components tested, free fatty acids produced the greatest cytotoxicity. The attraction and killing of leukocytes by comedonal components may be the mechanisms for the initiation or the enhancement (or both) of inflammation in acne vulgaris.

Concurrent linear scleroderma and systemic lupus erythematosus: a report of two cases.

Two patients with linear scleroderma (en coup de sabre) developed systemic lupus erythematosus (SLE). This association has been well documented in only one previous case. The presence of high titer antibodies to ribonucleoprotein (RNP) initially led to the diagnosis of the mixed connective tissue disease. Development of more serious clinical involvement and antibodies to Sm (case 1) or native deoxyribonucleic acid (nDNA) (case 2) helped establish a diagnosis of SLE. Use of these studies in the differential diagnosis of systemic rheumatic diseases is disucssed briefly. The presence of anti-RNAP antibodies in patients with localized scleroderma may herald a more serious rheumatic disease.

Expression of pemphigus vulgaris antigen in cultured human keratinocytes: effect of inhibitors, tunicamycin, and lectins.

We have investigated expression of pemphigus vulgaris antigen(s) in cultured human keratinocytes induced by the addition of extracellular calcium. Cycloheximide (10(-4) M) inhibited pemphigus antigen expression and stratification but actinomycin D (2 micrograms/ml) had no effect. Tunicamycin, which inhibits dolichol pyrophosphate-mediated glycosylation of asparaginyl residues specifically, was used to study the role of glycosylation. When calcium switching was carried out in the presence of tunicamycin, human keratinocytes did not stratify, and the expression of pemphigus antigen was partially inhibited and limited to cell-cell contact areas. Analysis of biosynthetically labeled proteins showed that the synthesis of high-molecular-weight proteins was markedly reduced in the tunicamycin-treated cells. A reciprocal blocking test demonstrated that concanavalin A and wheat germ agglutinin receptor share an epitope with pemphigus vulgaris antigen(s). These results suggest that Ca++, newly synthesized protein, and N-asparaginyl glycosylation are required for normal pemphigus antigen expression and epidermal stratification in vitro. Pemphigus vulgaris antigen may have a highly glycosylated, high-molecular-weight protein chain with carbohydrates playing an important role in epidermal cell morphology, adhesion, and stability of cell surface antigens.

Separation of epidermis from dermis with sodium thiocyanate.

After human skin is treated with 2 N sodium thiocyanate, epidermis is easily separated from dermis. The level of cleavage occurs at the lamina lucida of the basement membrane zone. Bullous pemphigoid antigen remains attached to the epidermis.

Immunopathology of cicatricial pemphigoid: studies of complement deposition.

Immunopathologic investigations were conducted on the sera and oral mucosal tissue specimens of 23 patients with cicatricial pemphigoid. A linear, continuous basement membrane zone pattern was noted in 83% of oral mucosal biopsy specimens studied. This pattern is indistinguishable from the pattern noted in immunofluorescence studies of bullous pemphigoid, herpes gestationis, and some cases of desquamative gingivitis. Complement studies provided data supportive of classical pathway activation in cicatricial pemphigoid tissue. Deposition of IgA with Factor B, properdin, and C3 raised the possibility of alternative pathway activation, a question requiring further study. Circulating antibasement membrane zone antibodies were noted in the sera of two patients with cicatricial pemphigoid.

Vasculitis in chronic urticaria: an immunopathologic study.

Forty-five patients with chronic urticaria were studied to determine: (1) the histologic incidence of leukocytoclastic vasculitis and (2) the clinical, laboratory and immunopathologic parameters that characterized this patient group. By histopathologic examination a spectrum of changes were noted as 9 patients showed leukocytoclastic vasculitis, 15 a dense perivascular infiltrate of lymphocytes and eosinophils, and 21 only a sparse lymphocytic perivascular infiltrate. Both the vasculitis and the dense infiltrate groups had an increased incidence of circulating immune complexes, as detected by Clq binding and monoclonal rheumatoid factor inhibition radioassays. Direct immunofluorescence showed blood vessel deposition of immunoglobulins, complement, and/or fibrin in 33% of the vasculitis group, 13% of the dense infiltrate group, and 9% of the sparse infiltrate group. These studies suggest that a meaningful number of patients with chronic urticaria have histologic and immunopathologic findings of vasculitis.

Recombinant desmoglein 3 has the necessary epitopes to adsorb and induce blister-causing antibodies.

The development of an animal model for studying the pathogenesis of pemphigus vulgaris (PV) has been hampered by the unavailability of the purified full-length autoantigen desmoglein 3 (Dsg 3).Therefore, we expressed Dsg 3 using a baculovirus expressed system. The expressed protein was identified as Dgs 3 by its reactivity with a pan-cadherin anti-serum, an anti-serum to a Dsg 3 synthetic peptide, or patient serum, and by amino-terminal sequencing. Carbohydrate analysis showed that recombinant Dsg 3 was glycosylated. While a majority of the recombinant protein was cell associated, by immunoprecipitation, some Dsg 3 was demonstrated in the medium. The Dgs 3 could adsorb out blister-causing antibodies from patient sera. Rabbit anti- Dsg 3 antibodies induced by the recombinant Dsg 3 showed specific binding to intercellular spaces of monkeys esophagus by indirect immunofluorescence. Moreover, these antibodies induced PV-like blisters in neonatal mice and weakly bound perilesional epidermis. Availability of large quantities of relatively pure Dsg 3 should now facilitate studies aimed at understanding Dsg 3 structure and pathogenesis of PV, with implications for developing specific immunotherapies.