Ramzy S. Labib

Paraneoplastic pemphigus. An autoimmune mucocutaneous disease associated with neoplasia.

Paraneoplastic pemphigus is a newly recognized disease that occurs in some patients with lymphoproliferative neoplasms and occasionally, solid tumors. Patients present with an acute illness of the mucosa and skin that shares clinical and histologic features with erythema multiforme, toxic epidermal necrolysis, and pemphigus vulgaris. These patients have antibodies against a complex of epithelial proteins that are present in desmosomes and hemidesmosomes. The course is usually fatal, except in some patients who undergo total resection of their neoplasm.

Induction of pemphigus in neonatal mice by passive transfer of igg from patients with the disease.

We examined the role of circulating autoantibodies in the pathogenesis of pemphigus vulgaris by passively transferring IgG fractions from five patients with pemphigus vulgaris into neonatal Balb/c mice, in doses of 1.5 to 16 mg per gram of body weight per day. Cutaneous blisters and erosions with the histologic, ultrastructural, and immunofluorescence features of pemphigus occurred in 39 to 55 mice given intraperitoneal injections of IgG from patients with pemphigus and in none of 58 control mice given normal human IgG. IgG fractions with high titers of pemphigus antibodies were most effective in inducing disease, and this effect was dose dependent. Titers of circulating IgG in mouse serum closely correlated with the extent of disease induced (P less than 0.002). This study strongly supports the proposed role of pemphigus autoantibodies in the pathogenesis of pemphigus vulgaris in human beings and demonstrates that pemphigus can be passively transferred to laboratory animals.

The Pathogenic Effect of IgG4 Autoantibodies in Endemic Pemphigus Foliaceus (Fogo Selvagem)

Endemic pemphigus foliaceus, or fogo selvagem, is an autoimmune blistering skin disease caused by IgG autoantibodies to a desmosome-associated glycoprotein. We studied the IgG subclasses with autoantibody activity in serum from 29 patients with active disease and in the skin lesions of 18 patients by immunofluorescence, using IgG-subclass-specific monoclonal antibodies. The predominant disease autoantibodies present in all patients were of the IgG4 subclass. IgG1 and IgG2 autoantibodies were detected in low titer in the 29 patients: IgG1 in 23 patients and IgG2 in 9. IgG3 autoantibodies were not detected in the serum of any patient. Direct immunofluorescence testing of skin lesions showed a preferential deposition of IgG4 on the keratinocyte surface. The pathogenic effect of IgG4 was demonstrated by the passive transfer of fractions containing IgG4 autoantibodies from the patients to neonatal BALB/c mice. The disease of the patients was reproduced clinically, histologically, and immunologically in these animals. Only IgG4 autoantibodies were detected by direct immunofluorescence, bound to the epidermis in the lesions of the mice, and by immunoelectron microscopy at the keratinocyte surface. IgG4 has previously been reported to be a blocking or protective antibody because it has poor effector functions in vitro, as compared with the other IgG subclasses. The finding that it is the pathogenic autoantibody in fogo selvagem raises the possibility that it may also be important in other autoimmune disease.

Herpes gestationis autoantibodies recognize a 180-kD human epidermal antigen.

Herpes gestationis (HG) is a putative autoimmune bullous dermatosis of pregnancy which shares many findings with bullous pemphigoid (BP), a disease of the elderly. This study identifies for the first time the antigen detected by HG autoantibodies and compares it with that recognized by BP autoantibodies. Sera from 16 HG and 17 BP patients, and from normal pregnant women were evaluated by immunofluorescent (IF) studies and immunoblotted against human epidermal extracts. 89% of HG sera with circulating antibodies by IF recognized a 180-kD protein by immunoblotting. 71% of BP sera recognized a 240-kD band, but 47% detected a 180-kD protein that comigrated with the antigen detected by HG sera. None of the control sera recognized any specific bands. These findings suggest that the 180-kD epidermal protein may be the antigen detected by the HG factor and they also define immunologic similarities between HG and BP.

Pathogenic effects of bullous pemphigoid autoantibodies on rabbit corneal epithelium.

Bullous pemphigoid (BP) is associated with circulating autoantibodies reactive with an antigen(s) of the basement membrane zone (BMZ) of skin and mucosae. The pathogenicity of these autoantibodies, although suspected, is unconfirmed. We have investigated the effects of BP autoantibodies on a closely related tissue, the corneal epithelium of the rabbit. IgG fractions from the sera of seven patients with BP were purified by (a) ammonium sulfate precipitation, (b) ion exchange chromatography, or (c) gel filtration. Control IgG was prepared by ion exchange chromatography of pooled normal human gamma globulins. 32 rabbits received corneal intrastromal injections of BP IgG fractions (50 microliter, 0.95-2.05 mg total dose) in one eye, and control IgG (50 microliter, 1.8 mg) in the contralateral cornea. 28 of 32 BP IgG injections produced corneal inflammatory lesions, 10 of which developed visible blisters. Histologically, lesions showed polymorphonuclear cells clustering along the BMZ, and subepithelial blister formation. Immunofluorescence showed in vivo bound IgG and C3 at the BMZ. The intensity of inflammation was dose dependent and correlated often with in vitro complement fixation titers of the fractions. None of 32 corneas injected with control IgG became inflamed. BP IgG fractions injected intradermally into the ear skin of rabbits failed to produce inflammation. This may be due to slow clearance of IgG in the cornea, and optimal binding by the corneal epithelium. The intracorneal injections of BP IgG reproduce the clinical, histological, and immunological features of BP. This study provides evidence that BP autoantibodies are pathogenic.

Pemphigus foliaceus antigen: Characterization of an immunoreactive tryptic fragment from BALB c mouse epidermis recognized by all patients' sera and major autoantibody subclasses

The pemphigus foliaceus antigen (PF Ag) is a 160-kDa desmosomal core glycoprotein, desmoglein I. A 50-kDa soluble immunoreactive fragment of the PF Ag was recently prepared from trypsinized cornified cell envelope preparations by papain treatment (R.S. Labib et al. 1989, J. Invest. Dermatol. 93, 272-279). This papain fragment (pf-PF) is associated with upper cell layers of the epidermis and appears to be trypsin resistant in situ. The present work describes the preparation of another fragment by trysinization of the viable lower cells of the epidermis of neonatal BALB/c mice. This tryptic fragment (tf-PF) is a 45-kDa glycoprotein that is partially purified by concanavalin A affinity chromatography of the trypsinization medium. The partially purified tf-PF preparation is capable of completely blocking the indirect immunofluorescence of high titer PF sera. The tf-PF is immunoprecipitated by all PF sera tested (n = 19) and by the two major subclasses of PF autoantibodies, IgG1 and IgG4. Autoantibodies of both the predominant IgG4 and the less prevalent IgG1 subclasses precipitate the same tf-PF as demonstrated by a single compact spot of pI 5.5 by two-dimensional polyacrylamide gel electrophoresis. Chemical and immunological comparison of the tf-PF and pf-PF may explain why the acantholytic lesions of PF appear only in the upper epidermis, despite the presence of the PF Ag throughout all layers of the epidermis. The availability of these two soluble immunoreactive fragments of the PF Ag will be of great value for the further immunochemical characterization of the antigenic epitopes and their role in cell-cell adhesion.

Elevated thymosin alpha I levels in Brazilian pemphigus foliaceus.

Levels of thymosin alpha I in the sera of 37 patients with Brazilian pemphigus foliaceus (BPF) were measured using a competitive binding radioimmunoassay. The values were compared with 19 patients with other forms of pemphigus, 13 relatives of patients with BPF, 18 patients with other dermatological diseases, and 265 normal controls. We found that 27 (73%) of the patients with BPF had thymosin alpha I serum levels that were at least two standard deviations above the mean for normal individuals. The mean value for patients with BPF was significantly greater than any other groups studied. The thymosin elevation is similar to alterations seen in certain viral diseases and suggests that BPF is aetiopathogenically distinct from the other forms of pemphigus.

Selective surface radioiodination of keratinocytes in primary culture labels a 59 kD keratin and other surface proteins

SummaryMurine keratinocytes were vectorially labeled in order to identify and characterize surface proteins. Freshly trypsinized keratinocytes were obtained from neonatal BALB/c mice and primary cultures were established. Cells were radiolabeled with iodogen and 125I immediately after harvesting and after 4, 24, 48, and 72 h in culture. Cells were solubilized with 1% sodium dodecyl sulfate (SDS) and reducing agents (total extract), or sequentially extracted with: (a) 1% Triton X-100 (membrane/cytosol fraction); (b) 2 M NaCl (salt fraction); and (c) 1% sodium dodecyl sulfate (SDS; cytoskeleton fraction). Extracts were analyzed by 1- or 2-D polyacrylamide gel electrophoresis (PAGE), followed by autoradiography (AR). The 59 kD acidic murine keratin was identified by immunoblot analysis with monoclonal antibody AE1 and a human polyclonal anti-59 kD keratin autoantibody designated Cascas-42. Total extracts contained up to ten labeled proteins ranging from 10 to 180 kD. Eight of these proteins were present in the membrane/cytosol fraction (10, 12, 18, 30, 38, 41, 66, and 130 kD). Intense labeling of a 59 kD cytoskeletal protein was consistently seen; this protein was identified as the 59 kD acidic murine keratin (Molls catalogue no. 10) by immunoblot analysis of extracted proteins separated by 1-D and 2-D PAGE. A 180 kD protein had variable solubility characteristics, fractionating with either the cytoskeleton or membrane/cytosol as a function of time. No labeled proteins were detected in the salt extract, and neither actin nor other major cytoskeletal proteins were radiolabeled. These studies demonstrate the cell-surface disposition of at least ten keratinocyte proteins, and suggest that the 59 kD murine acidic keratin has a surface-exposed domain.

Epidermal proteins. I. Differential extraction and quantitative polyacrylamide gel-electrophoretic analysis of basal spinous-cell proteins of neonatal mouse epidermis

SummaryThe study of keratinocyte proteins and their changes in different physiological, experimental, and pathological states has been facilitated and stimulated by the development of high-resolution polyacrylamide gel electrophoretic (PAGE) techniques. We describe a differential extraction system that separates the keratinocyte proteins into four major groups which are further quantitatively analyzed by PAGE: (1) cytoplasmic-soluble proteins, (2) nonionic-detergentsoluble proteins consisting of membrane-associated protein, (3) salt-dissociated proteins mainly consisting of histones, and ribosomal and keratohyaline granule proteins, and (4) the keratins (and other intermediate-filament-associated proteins), which are further separated into disulfide-stabilized keratins and keratins that do not require reducing agents in order to dissolve in sodium dodecyl sulfate (SDS) or urea. This extraction system was applied to neonatal mouse epidermal cell preparations that consisted mainly (60%–85%) of basal cells and also of some spinous cells (10%–30%). The SDS-PAGE patterns obtained by spectrophotometric scanning were graphically compared and integrated using an IBM personal computer. The protein bands in each extract were identified by their apparent molecular weights and were quantitated as a percentage of the total in each extract and in micrograms per 15×106 cells. Some protein peaks were provisionally identified as actin, the core histones H2A, H2B, H3, and H4, ribosomal proteins, and six keratins. This study serves as the foundation for the quantitative description of molecular changes which occur during keratinocyte differentiation and for the comprehensive identification of epidermal proteins.

The pemphigus antigens

Abstract Pemphigus antigens are normal squamous epithelial tissue constituents that are identified by their reaction with autoantibodies found in the serum of patients with active pemphigus vulgaris or any of the other types of pemphigus ( i.e. , vegetans, erythematosus, and foliaceous). The characteristic lesion of pemphigus is epidermal suprabasilar cell-cell detachment (acantholysis). It is important to note that pemphigus antigens are present in the intercellular substance (ICS) of stratified squamous epithelia where this process of acantholysis occurs. There is increasing evidence that pemphigus antigens are heterogenous, but whether this heterogeneity is due to the existence of different antigen molecules or different antigenic determinants on the same molecule is not yet clear. The use of the word antigens in this article does not imply commitment or rejection of any of these alternative explanations.