A sensitive and specific assay for the serological diagnosis of antilaminin 332 mucous membrane pemphigoid.

BACKGROUND: Antilaminin 332 mucous membrane pemphigoid (MMP) is an autoimmune subepidermal blistering disease with predominant mucosal involvement and autoantibodies against laminin 332. Malignancies have been associated with this disease; however, no standardized detection system for antilaminin 332 serum antibodies is widely available. OBJECTIVES: Development of a sensitive and specific assay for the detection of antilaminin 332 antibodies. METHODS: An indirect immunofluorescence (IF) assay using recombinant laminin 332 was developed and probed with a large number of antilaminin 332 MMP patient sera (n = 93), as well as sera from patients with antilaminin 332-negative MMP (n = 153), bullous pemphigoid (n = 20), pemphigus vulgaris (n = 20) and noninflammatory dermatoses (n = 22), and healthy blood donors (n = 100). RESULTS: In the novel IF assay, sensitivities with the laminin 332 heterotrimer and the individual α3, ß3 and γ2 chains were 77%, 43%, 41% and 13%, respectively, with specificities of 100% for each substrate. The sensitivity for the heterotrimer increased when an anti-IgG4 enriched antitotal IgG conjugate was applied. Antilaminin 332 reactivity paralleled disease activity and was associated with malignancies in 25% of patients with antilaminin 332 MMP. CONCLUSIONS: The novel IF-based assay will facilitate the serological diagnosis of antilaminin 332 MMP and may help to identify patients at risk of a malignancy.

C5a stimulates secretion of tumor necrosis factor from human mononuclear cells in vitro. Comparison with secretion of interleukin 1 beta and interleukin 1 alpha.

We have demonstrated that purified C5a is a potent stimulus to human PBMC secretion of TNF-alpha, IL-1 beta, and IL-1 alpha, which proceeds in a dose-dependent fashion. At a given concentration of C5a, TNF-alpha and IL-1 beta secretion did not differ significantly; both were secreted in significantly greater quantity than IL-1 alpha. Clinical conditions such as Gram-positive and Gram-negative bacterial infections, trauma, and immune complex diseases activate complement. Through the mediation of TNF and IL-1 secreted in response to C5a, these diverse disorders can share common features of fever, coagulopathy, acute phase protein production, and disordered metabolism.

Passive transfer of anti-laminin 5 antibodies induces subepidermal blisters in neonatal mice.

Patients with a recently identified subepithelial blistering disease have IgG anti-laminin 5 autoantibodies. To determine if such antibodies can be pathogenic in vivo, we developed and characterized rabbit anti-laminin 5 IgG, and passively transferred these antibodies to neonatal mice. Immune rabbit IgG specifically bound human and murine epidermal basement membranes, immunoblotted and immunoprecipitated all laminin 5 subunits from extracts of human and murine keratinocytes, and showed no reactivity to other keratinocyte proteins or epithelial basement membranes that do not contain laminin 5. Mice (n = 29) receiving purified anti-laminin 5 IgG developed, in a dose-related fashion, circulating anti-laminin 5 antibodies, deposits of rabbit IgG and murine C3 in epidermal basement membranes, and subepidermal blisters of skin and mucous membranes. No alterations developed in controls (n = 14) receiving identical amounts of normal rabbit IgG. Passive transfer of anti-laminin 5 (but not control) IgG to neonatal C5- (n = 3) or mast cell-deficient (n = 3) mice produced subepidermal blisters with the same clinical, histologic, and immunopathologic features as those documented in BALB/c mice. These studies establish an animal model of a human blistering disease that can be used to define disease mechanisms and treatment modalities.

Revertant mosaicism: partial correction of a germ-line mutation in COL17A1 by a frame-restoring mutation.

Generalized atrophic benign epidermolysis bullosa is an autosomal recessive subepidermal blistering disease typified by null mutations in COL17A1. In 1 large kindred, affected individuals were homozygous for a 2-bp deletion in COL17A1, 4003delTC, which resulted in a downstream premature termination codon, nonsense-mediated mRNA decay, and abrogation of type XVII collagen synthesis. Interestingly, 1 of these patients, although phenotypically identical to her affected siblings, showed focal expression of type XVII collagen in epidermal basement membrane in a pattern suggestive of revertant mosaicism. When studies of randomly obtained epidermal, oromucosal, and peripheral blood cells failed to identify the genetic basis of this apparent mosaicism, microscopic subpopulations of potentially revertant epidermal cells (i.e., those overlying basement membrane containing type XVII collagen) were selectively isolated using laser capture microdissection. Analysis of DNA and RNA from these cells revealed a second mutation, 4080insGG, on 1 allele of COL17A1. This 2-bp insertion corrected the reading frame just proximal to the premature termination codon, countered nonsense-mediated mRNA decay, and allowed protein production by patient keratinocytes in vivo and in vitro. These studies elucidate the molecular basis of a novel form of revertant mosaicism in humans.

Epiligrin, the major human keratinocyte integrin ligand, is a target in both an acquired autoimmune and an inherited subepidermal blistering skin disease.

Epiligrin, the major component of human keratinocyte extracellular matrix, serves as the preferred integrin ligand for alpha 3 beta 1 in plasma membranes and focal adhesions, and colocalizes with alpha 6 beta 4 in hemidesmosomes. In human skin, epiligrin is found in the lamina lucida subregion of epidermal basement membrane, where it is thought to be associated with anchoring filaments. We have identified three patients with an acquired mucosal predominant subepidermal blistering disease who have IgG anti-basement membrane autoantibodies that bind the lamina lucida/lamina densa interface of epidermal basement membrane, stain cultured human keratinocyte extracellular matrix, and immunoprecipitate disulfide linked polypeptides of 170, 145, 125, and 95 kD in human keratinocyte culture media in a pattern identical to that of P1E1, a murine monoclonal antiepiligrin antibody. Comparative immunoprecipitation studies of patient sera, P1E1, and GB3 monoclonal antibody show that epiligrin is identical to the antigen (i.e., BM600 or GB3 antigen) previously reported to be absent from the skin of patients with lethal junctional epidermolysis bullosa, an inherited subepidermal blistering disease. Moreover, skin from a fetus with this disease shows no evidence of reactivity to patient antiepiligrin autoantibodies or P1E1. These studies show that antiepiligrin autoantibodies are a specific marker for a novel autoimmune blistering disease and that the epidermal basement membrane antigen absent in patients with lethal junctional epidermolysis bullosa is epiligrin.

Studies of human C5a as a mediator of inflammation in normal human skin.

C5a is an 11,000-D fragment of the fifth component of complement (C5) with potent anaphylatoxic and leukocyte chemotactic activities. C5a is believed to play an important role in the pathophysiology of certain skin disorders and systemic diseases with cutaneous manifestations. However, there is very little known about the in vivo reactivity of C5a in man. In this study we examined the effects of intradermal injections of human C5a in 17 normal volunteers. C5a was prepared by interacting highly purified human C5 with zymosan bound alternative pathway C5 convertase under conditions resulting in consumption of approximately 90% of the C5 substrate. C5a produced in this manner was chemotactic for human neutrophils and monocytes (0.5 X 10(-7) to 10(-9) M) and caused neutrophil aggregation and myeloperoxidase release (concentrations greater than or equal to 10(-10) M) in vitro. In vivo, C5a produced immediate wheal and flare reactions in all volunteers, and was active in doses as low as 1 ng (10(-13) mol). Intradermal testing with 20 ng of C5a in eight volunteers produced a maximal mean wheal of 11.75 mm (+/- 0.80 mm SEM) 20 min after anaphylatoxin injection, and a maximal mean erythema of 62.50 mm (+/- 3.27 mm SEM) 10 min after C5a administration. Reactions at C5a test sites were dose-related, associated with marked pruritus in some subjects, resolved without lesion formation, and were not associated with late phase reactions. In vivo testing revealed that human C5a was a more potent mediator of wheal and flare reactions than histamine, 48/80, human C3a, or morphine sulfate. Skin biopsies from eight volunteers 20 min after intradermal injection of 20 ng of C5a revealed a neutrophil-predominant perivascular infiltrate, endothelial cell edema, and sites of leukocytoclasis. Mast cell degranulation was observed on both light and electron microscopy of biopsies from C5a test sites. Although erythema at C5a injection sites was reduced by pretreating volunteers with hydroxyzine, whealing reactions and cellular infiltrates in biopsies were unaffected by this H1-antihistamine. Moderate doses of systemic corticosteroids did not alter clinical or histologic reactions at C5a injection sites in two volunteers. This study, using doses within the potential physiologic range of the anaphylatoxin, provides a comprehensive assessment of the effect of human C5a on normal human skin.

Role of C5a in the induction of tumoricidal activity in C3H/HeJ (Lpsd) and C3H/OuJ (Lpsn) macrophages.

Thioglycollate-elicited macrophages from C3H/HeJ (Lpsd) and C3H/OuJ (Lpsn) mice were cultured in a two-signal, tumoricidal assay using recombinant interferon-gamma (rIFN-gamma) as the "priming" signal and recombinant human C5a (rC5a) as the "trigger" signal. These experiments were compared directly with a well established, two-signal tumoricidal assay in which rIFN-gamma was used as the "priming" signal and protein-rich, butanol-extracted lipopolysaccharide (But-LPS) as the "trigger" signal. These studies showed that rIFN-gamma-primed macrophages can be triggered in a dose-dependent manner by rC5a to effect high levels of tumoricidal activity. Maximum levels of cytotoxicity achieved using this endogenously produced, biologically active peptide as a "trigger" signal were comparable to those obtained using But-LPS. Moreover, experiments in which anti-C5 antibody was included in macrophage cultures stimulated with rIFN-gamma and But-LPS showed a significant reduction (P less than .05) in tumoricidal activity. Because LPS has been shown to induce macrophage C5 production and enzyme release, these findings suggest that macrophage-derived C5 is locally converted to C5a (or some other biologically active C5 cleavage fragment), which functions as an autocrine trigger signal for the induction of tumoricidal activity. In summary, these data suggest 1) that rC5a can provide a "second signal" to rIFN-gamma-primed murine macrophages for the induction of tumoricidal activity and 2) that macrophage-derived C5 or C5a may represent an autocrine signal induced by exogenous "trigger signals."

Differential release of mediators from human basophils: differences in arachidonic acid metabolism following activation by unrelated stimuli.

We have examined the release of histamine and LTC4 from purified human basophils challenged with several different stimuli, both physiological and nonphysiological. Basophils (n = 16) challenged with 0.1 micrograms/ml anti-IgE released 38 +/- 4% of their available histamine and 39 +/- 12 ng LTC4/10(6) basophils within 15-30 min. F-Met peptide (n = 8) caused the release of 54 +/- 8% histamine and 42 +/- 25 ng LTC4/10(6) basophils within a period of 2-5 min. C5a caused the release of 22 +/- 3% histamine from selected donors but failed to initiate any LTC4 release unless combined with D2O or 5 mM extracellular calcium. The two nonphysiological stimuli A23187 and TPA caused extensive histamine release, 67 +/- 8 and 82 +/- 11%, respectively, and while A23187 initiated a large and rapid release of leukotriene, TPA failed to release any LTC4 even when combined with D2O or 2-5 mM extracellular calcium. Increased concentrations of extracellular calcium enhanced anti-IgE and f-Met peptide induced release of LTC4 but inhibited the A23187 induced release of leukotriene. A single peak of immunoreactive leukotriene C4 that comigrated with the authentic standard was identified using HPLC followed by radioimmunoassay. No LTD4 or LTE4 could be detected. Purified human basophils incubated with 0.2 microM [3H]AA incorporated 290 pmol/10(6) cells, or 32 +/- 5% of the available label within 60 min. The [3H]AA was taken principally into the phospholipids (73 +/- 5%), with 20 +/- 3% as neutral lipid, and only 5 +/- 2% remaining as the free acid. Three phospholipid subclasses, phosphatidylcholine, PC (24 +/- 2%), phosphatidylinositol, PI (22 +/- 1%), and phosphatidylethanolamine, PE (15 +/- 3%), accounted for the majority of the incorporated [3H]AA while the remainder of the phospholipids accounted for less than 5% of the total cpm. HPLC analysis of the lipid mediators released during stimulation with 0.1 micrograms/ml anti-IgE revealed [3H]LTC4 (2.4 +/- 1.0%), [3H]5HETE (1.0 +/- 0.1%), unmetabolized [3H]AA (91 +/- 2%), and an unidentified peak (3.4 +/- 1.4%). The unknown metabolite eluted with the prostaglandins, was inhibited by indomethacin, and appeared to have a relatively high specific activity. It may thus represent an artifact of the labeling procedure rather than a novel basophil-derived prostaglandin.

Studies of C3d,g in normal human epidermal basement membrane.

Recent studies in our laboratory have demonstrated that polyclonal as well as monoclonal antibodies directed against C3d,g, a specific 41,000-Da fragment of the third component of complement, bind normal human epidermal basement membrane in a continuous pattern. No such reactivity is present within dermal microvascular basement membranes. By direct immunofluorescence microscopy, anti-human C3d,g antibodies bind the base of the cleavage plane in 1 M NaCl split human skin. By immunoelectron microscopy, anti-human C3d,g reactivity is found along the base of the lamina densa and in the sublamina densa region. Control antibodies directed against human C3, C3c, C5, IgG, IgA, or IgM do not bind normal human epidermal basement membrane or identify in situ deposits of immune complexes in multiple samples of normal human skin that are all positive for C3d,g. Preabsorption of monoclonal or polyclonal anti-human C3d,g antibodies with purified human C3d completely blocks these reagents' epidermal basement membrane reactivity. Studies of skin samples from a patient with congenital C3 deficiency reveal that anti-human C3d,g antibodies do not bind this subject's epidermal basement membrane. Moreover, in vitro treatment of this patient's skin with normal serum, aged serum containing C3d,g, purified human C3, or zymosan-serum reaction mixtures does not restore epidermal basement membrane anti-human C3d,g binding. Studies of other primate tissues demonstrate that C3d,g is not restricted to basement membranes of stratified squamous epithelia as it is also present within renal tubule and glomerular basement membranes. While a recent study has demonstrated that C3d binds laminin in vitro, our investigations show a difference in both the regional and ultrastructural distribution of laminin and C3d,g in normal human skin. Furthermore, C3d,g is absent from laminin-rich basement membranes of papulonodular basal cell carcinomas. These findings suggest that C3d,g is not passively incorporated within selected epithelial basement membranes but rather is a previously unrecognized normal constituent. Basement membrane-associated C3d,g may play a role in adhesive interactions between leukocytes and matrix proteins. Moreover, a C3d,g binding site(s) in selected epithelial basement membranes may account for the accumulation of C3 containing immune complexes in such tissues.

Inflammatory properties of human C5a and C5a des Arg/ in mast cell-depleted human skin.

C5a and its degradation product, C5a des Arg, elicit immediate cutaneous inflammatory reactions after intradermal injection. Histologically, these reactions are characterized by neutrophil-rich leukocytic infiltrates, leukocytoclasis, edema, and dermal mast cell degranulation. It has not been possible to assess in vivo the relative contributions of resident mast cells and circulating leukocytes to this reaction because the accumulation of leukocytes and degranulation of mast cells occur simultaneously after injection of these anaphylatoxins. To assess the role of mast cells in these inflammatory reactions, we have examined the reactivity of human skin selectively depleted of dermal mast cells by local corticosteroid treatment. Corticosteroid-treated skin became virtually devoid of dermal mast cells within 4-6 wk as assessed by light microscopy, immunofluorescence with fluorescein-conjugated avidin, or electron microscopy. Mast cell-depleted skin demonstrated normal vasopermeability and vasodilatory responsiveness to intradermal injection of histamine, but the reactivity of these sites to the mast cell secretagogue, morphine, was absent. Moreover, no clinical reactions were detectable in mast cell-depleted human skin after intradermal challenge with 50 ng of either C5a or C5a des Arg, despite the fact that biopsies of these sites revealed substantial, neutrophil-rich infiltrates. These infiltrates were qualitatively and quantitatively identical to C5a or C5a des Arg-induced infiltrates in mast cell replete skin. This experimental approach in vivo has allowed the independent analysis of the anaphylactogenic and chemoattractant activities of human C5a and C5a des Arg in human skin, demonstrated the importance of dermal mast cells in these clinical responses, and shown that leukocytes can accumulate at these injection sites directly in response to these mediators.

Anti-basement membrane autoantibodies in patients with anti-epiligrin cicatricial pemphigoid bind the alpha subunit of laminin 5.

Recent studies have identified a group of cicatricial pemphigoid patients who have IgG anti-basement membrane autoantibodies that recognize epiligrin, a set of disulfide-linked polypeptides closely related if not identical to laminin 5 (formerly called kalinin, nicein, or BM600). To further understand the pathophysiology of blister formation in these patients, we have sought to identify the specific polypeptide(s) targeted by their autoantibodies. Comparative studies show that sera from these patients (nine of nine), P1E1 monoclonal anti-epiligrin antibody, and polyclonal as well as monoclonal anti-laminin 5 antibodies immunoprecipitate the same set of disulfide-linked polypeptides from media of biosynthetically radiolabeled human keratinocytes. Moreover, sera from eight of nine patients with anti-epiligrin cicatricial pemphigoid immunoblot the alpha subunit of laminin 5 but show no reactivity to its beta or gamma subunits. In addition, circulating IgG from a representative patient was affinity-purified against the alpha subunit of laminin 5 and shown to bind the dermal side of 1 M NaC1 split skin in the same manner as autoantibodies from all patients with anti-epiligrin cicatricial pemphigoid. Sera from patients with bullous pemphigoid (n = 5), other forms of cicatricial pemphigoid (n = 5), epidermolysis bullosa acquisita (n = 4), or bullous systemic lupus erythematosus (n = 1) show no reactivity against any subunit of this laminin isoform in immunoprecipitation or immunoblot experiments. These findings correlate with prior reports showing that a monoclonal antibody directed against the alpha subunit of laminin 5 (i.e., laminin subunit alpha 3) induces detachment of human keratinocytes from extracellular matrix in vitro as well as epidermis from human skin in situ. Together, these studies suggest that laminin subunit alpha 3 mediates attachment of basal keratinocytes to epidermal basement membrane and that autoantibodies directed against it may be pathogenic.

Analysis of the interaction of human C5a and C5a des Arg with human monocytes and neutrophils: flow cytometric and chemotaxis studies.

C5a and C5a des Arg are potent complement-derived mediators that bind receptors on peripheral blood leukocytes and tissue-specific cellular elements to elicit and amplify inflammatory and immunomodulatory reactions. To study the interactions of C5a and C5a des Arg with these cells, fluorescein conjugates of these ligands were prepared by a new technique and their binding to monocytes, neutrophils, platelets, and endothelial cells was studied with flow cytometry. Fluoresceinated C5a produced neutrophil myeloperoxidase release and chemotaxis and also bound rabbit anti-C5a antibody much like native anaphylatoxin; likewise, fluoresceinated C5a des Arg demonstrated retention of biologic and antigenic activities. Both fluorescein-conjugated C5a and C5a des Arg bound to monocytes and neutrophils in a concentration-dependent, saturable, and homogeneous manner, but 10- to 15-fold higher concentrations of C5a des Arg were required to attain saturable binding of these leukocytes. Ligand binding was specifically inhibited by native purified human C5a in a concentration-dependent manner, while it was unaffected by C3a or N-formyl-methionyl-leucyl-phenylalanine-lysine. There was no evidence of a C5a receptor-negative subpopulation of monocytes or neutrophils. Moreover, comparative binding experiments with leukocytes from multiple normal volunteers showed that a greater percentage of monocytes than neutrophils bound C5a at less than saturable concentrations of ligand (P less than 0.05, 0.5 to 5.0 nM). A representative half-maximal binding of fluorescein-conjugated C5a (C5a des Arg) binding to monocytes and neutrophils was 1.2 nM (30 nM) and 2.6 nM (68 nM), respectively. In contrast, fluorescein-conjugated C5a did not specifically bind to human platelets or umbilical vein endothelial cells.

Noncomplement fixing, IgG4 autoantibodies predominate in patients with anti-epiligrin cicatricial pemphigoid.

This study characterized the specific reactivity, IgG subclass, and complement fixing ability of anti-laminin-5 IgG from 12 patients with anti-epiligrin cicatricial pemphigoid. Circulating IgG from all patients bound the dermal side of 1 M NaCl split skin, immunoprecipitated laminin-5 produced by biosynthetically radiolabeled human keratinocytes, and (in 10 of 12 cases) immunoblotted the laminin-alpha3 subunit. Analysis of the distribution of IgG subclasses in these patients' circulating anti-laminin-5 autoantibodies by semiquantitative indirect immunofluorescence microscopy using the HP series of subclass-specific monoclonal antibodies revealed: (i) IgG4 predominant autoantibodies in seven of 11 sera; (ii) IgG1 and IgG2 at substantially lower levels in a smaller number of sera; and (iii) no specific IgG3 anti-laminin-5 autoantibodies in any patients. The same IgG4-dominant profile of anti-laminin-5 autoantibodies was found in enzyme-linked immunosorbent assay studies of purified human laminin 5. Direct immunofluorescence microscopy of six skin biopsies from three patients found that IgG4 was also the predominant subclass of IgG in epidermal basement membranes in situ. Consistent with these findings, sera from 11 of 11 patients with anti-laminin-5 IgG autoantibodies did not fix C3 to epidermal basement membranes in vitro. These immunochemical studies suggest that complement activation does not play a major role in the pathophysiology of this disease and that subepidermal blisters in these patients may develop via a direct effect of anti-laminin-5 IgG itself.

Human anti-laminin 5 autoantibodies induce subepidermal blisters in an experimental human skin graft model.

Patients with one form of cicatricial pemphigoid have IgG antibasement membrane autoantibodies against laminin 5 (alpha3beta3gamma2). Although passive transfer of rabbit anti-laminin 5 IgG to neonatal mice has been shown to induce subepidermal blisters that mimic those in patients, it has not been possible to directly assess the pathogenic activity of human autoantibodies in this animal model because the latter do not bind murine skin. To address this question, a disease model in adult mice as well as SCID mice bearing human skin grafts was developed. Adult BALB/C mice challenged with rabbit anti-laminin 5 IgG developed, in a concentration-related fashion, erythema, erosions, and crusts surrounding injection sites, histologic evidence of noninflammatory, subepidermal blisters, and deposits of rabbit IgG and murine C3 in epidermal basement membranes. Anti-laminin 5 IgG also induced subepidermal blisters in: adult complement-, mast cell-, and immuno-deficient mice; adult BALB/C mice pretreated with dexamethasone; and human skin grafts on SCID mice. Alterations did not develop in matching controls challenged with identical amounts of purified normal rabbit IgG or bovine serum albumin. Using this adult mouse model, human skin grafts on SCID mice were challenged with purified IgG from patients with alpha subunit-specific, anti-laminin 5 autoantibodies, or normal controls. Patient (but not control) IgG induced epidermal fragility as well as noninflammatory, subepidermal blisters in grafted human (but not adjacent murine) skin. Moreover, whereas all mice that received patient autoantibodies had anti-laminin 5 IgG in their circulation, deposits of human IgG were present only in the epidermal basement membranes of grafts. Interestingly, these in situ and circulating autoantibodies were predominately of the IgG4 subclass. These studies demonstrate that human anti-laminin 5 autoantibodies are pathogenic in vivo and describe an animal model that can be used to define disease pathomechanisms and biologically important domains within this autoantigen.

A-431 cells and human keratinocytes synthesize and secrete the third component of complement.

Biosynthetic radiolabeling studies demonstrate that A-431 cells, a human epidermoid carcinoma cell line, and human keratinocytes synthesize and secrete C3 as two disulfide-linked polypeptide chains of 120 and 75 kD. Moreover, epithelial cell-derived C3 co-migrates in SDS-PAGE with that produced by HepG2 cells, a human hepatoma cell line previously used to elucidate complement component biosynthesis. Pulse-chase studies in A-431 cells demonstrate that epithelial cell-derived C3 is produced as a 195-kD precursor molecule, pro-C3, which is processed intracellularly by limited proteolysis into 120- and 75-kD C3 alpha and beta chains. Comparative studies demonstrate that A-431 cell-derived C3 is synthesized, processed, and secreted in parallel but in lower quantity than that produced by HepG2 cells. Treatment of biosynthetically labeled A-431 cell culture supernatants with normal human serum and zymosan produces C3 alpha chain cleavage and specific C3 fragments that are not present in control culture supernatants treated with heat-inactivated human serum and zymosan. Northern blot analysis of total cellular RNA extracted from A-431 cells, human keratinocytes, and HepG2 cells reveals quantitative identity of a 5.1-kb C3 mRNA species in these three cell types. Epithelial cell-derived C3 may play an important role in local inflammatory and immunologic reactions including such reactions in human skin. Moreover, epithelial cell C3 synthesis may have direct relevance to the recent demonstration of C3d,g within selected normal primate epithelial basement membranes, including epidermal basement membrane.

Patients with bone marrow failure demonstrate decreased cutaneous reactivity to human C5a.

In vivo studies have shown that human C5a, a potent complement-derived anaphylatoxin and chemoattractant, produces immediate inflammatory reactions following intradermal injection in human skin. At concentrations within its potential physiologic range, intradermal injection of C5a elicits immediate wheal and flare reactions, increased vascular permeability, mast cell degranulation, and neutrophil-rich infiltrates. To assess the relative contribution of interacting cellular elements to C5a-induced inflammation in normal human skin, purified human C5a was tested intradermally in 8 patients with bone marrow failure (BMF). Reactions to C5a in patients with BMF were compared with responses at identical test sites in healthy volunteers and other patients with cutaneous disorders. Patients with BMF demonstrated significantly less wheal and flare reactivity following intradermal injection of C5a than controls (p less than 0.05 and less than 0.02, respectively). In these studies, patients with the greatest cytopenia generally showed the least cutaneous reactivity to human C5a. Biopsies of C5a test sites in patients with BMF revealed an absence of leukocytes in marked contrast to neutrophil-rich infiltrates observed at test sites in healthy volunteers. Avidin-fluorescent and/or Giemsa staining of skin biopsies revealed no difference between the number of dermal mast cells in patients with BMF and samples of normal human skin. In addition, skin test studies with histamine (2 micrograms) and morphine (5 micrograms) performed to assess cutaneous vascular and mast cell responsiveness in patients with BMF, normal volunteers and controls with rhinitis revealed no significant differences in cutaneous reactivity to these pharmacologic agents. These in vivo studies demonstrate that patients with BMF specifically exhibit decreased cutaneous reactivity to human C5a and suggest that neutrophils make an important and an immediate contribution to inflammatory responses elicited by this anaphylatoxin.

Human keratinocytes and A-431 cells synthesize and secrete factor B, the major zymogen protease of the alternative complement pathway.

Biosynthetic radiolabeling studies demonstrate that human keratinocytes and A-431 cells, a human epidermoid carcinoma cell line, synthesize and secrete factor B as a monomeric 105-kD protein. Epithelial cell-derived factor B comigrates in SDS-PAGE with that produced by HepG2 cells, a human hepatoma cell line traditionally employed in studies of complement component biosynthesis. Comparative pulse-chase studies in A-431 and HepG2 cells show that this alternative pathway complement component is produced as co-migrating 100-kD intracellular proteins that are processed in both cell types to 105-kD extracellular factor B. Quantitatively, immunoprecipitable factor B accounts for 0.05% of radiolabeled proteins in A-431 cell culture media. Treatment of biosynthetically radiolabeled A-431 cell culture media with cobra venom factor and factor D for 60 min at 37 degrees C produces the specific factor B cleavage products Ba and Bb. These fragments are not identifiable in control culture media subjected to similar treatment in the absence of alternative pathway activators. Northern blot analysis of total cellular RNA from human keratinocytes, A-431 cells, and HepG2 cells reveals qualitative identity of a 2.8-kb factor B mRNA species in these three cell types. The relative level of factor B mRNA expression in these cells parallels their level of factor B protein synthesis (i.e., HepG2 cells greater than A-431 cells greater than human keratinocytes). Epithelial cell-derived factor B may play an important role in local inflammatory reactions and also directly interact with epithelial cell derived C3--a key classical and alternative pathway complement component recently shown to be produced by human keratinocytes.

The extracellular domain of BPAG2 localizes to anchoring filaments and its carboxyl terminus extends to the lamina densa of normal human epidermal basement membrane.

Bullous pemphigoid antigen 2 (BPAG2) is a 180 kDa type II transmembrane protein associated with hemidesmosomes (HDs) in basal keratinocytes. To better understand how BPAG2 promotes keratinocyte adhesion to epidermal basement membrane (BM), purified IgG against a baculovirus-encoded recombinant was used to localize its carboxyl terminus in human skin by immunogold electron microscopy (IEM). A 2.1-kb BPAG2 cDNA encoding the distal extracellular domain and carboxyl terminus of BPAG2 was used in a baculovirus expression system to create virus that produced a 70-kDa recombinant form of BPAG2 (BV4). BV4 was purified, characterized, and used to raise high-titer specific rabbit IgG. Purified anti-BV4 IgG bound the epidermal side of 1 M NaCl split skin and bound only BPAG2 on immunoblots containing extracts of human keratinocytes. In IEM studies of pre- and post-embedded skin, the distal ectodomain of BPAG2 localized beneath HDs in basal keratinocytes; there was no evidence of BPAG2 beneath melanocytes. Anti-BV4 IgG extensively bound anchoring filaments on the epidermal side of 1 M NaCl split skin; this staining extended along anchoring filaments to their ends. In post-embedded skin, the carboxyl terminus of BPAG2 was localized within the lamina densa, 41 nm (mean of 400 determinations) beneath plasma membranes of basal keratinocytes. BPAG2 thus extends from the intracellular HD plaque of basal keratinocytes to the lamina densa of human epidermal BM.

Dietary gluten challenge does not influence the levels of circulating immune complexes in patients with dermatitis herpetiformis.

To examine the relationship between the gluten-sensitive enteropathy (GSE) and IgA circulating immune complexes (CIC) in dermatitis herpetiformis (DH) a series of dietary gluten-challenge studies were performed in patients with DH and patients with ordinary GSE. Serial serum samples were monitored for IgA-, IgG-, and IgM-containing CIC levels. In the first study, 9 DH patients and 5 controls were fed 20 g of gluten flour as a breakfast meal on 1 of 2 consecutive study days. DH patients did not develop or increase their levels of CIC after gluten-challenge or gluten-free meals. There was no significant difference between the DH patients and the control group in regard to development of CIC. To evaluate the effect of dietary gluten in another form, 8 DH patients were given meals containing 100 g of boiled Canadian cracked wheat. Two patients with ordinary GSE were also challenged with cracked wheat. Again there was no elevation or induction of CIC above baseline determinations by gluten-challenge meals. These studies suggest that dietary gluten does not induce the formation of CIC in patients with DH.