Susan A. Krasny

Sensitivity and specificity of IgA-class antiendomysial antibodies for dermatitis herpetiformis and findings relevant to their pathogenic significance

The specificity and sensitivity of the recently reported IgA-class antiendomysial antibody test for gluten-sensitive enteropathy were evaluated in four double-blind studies involving the sera of fifty-seven patients with dermatitis herpetiformis who were not on a gluten-free diet and ninety-seven assorted control sera. The control sera provided by the four centers included the sera of nineteen patients with dermatitis herpetiformis and two with celiac disease who were on a gluten-free diet, the sera of five normal subjects with human lymphocyte antigens (B8 locus), the sera of thirteen patients with linear IgA bullous dermatosis, and fifty-eight other control sera, mostly from patients with other bullous diseases and other dermatoses. The frequency of IgA antiendomysial antibody in these coded studies was zero of ninety-seven control sera and thirty-four of fifty-seven sera (60%) from patients with dermatitis herpetiformis who were not on a gluten-free diet. The pathogenic role of IgA antiendomysial antibodies in dermatitis herpetiformis and celiac disease is suggested not only by their high degree of disease sensitivity and specificity but also by their formation in response to gluten challenge, their appearance before gut changes, and the in vitro binding of gliadin to the antiendomysial antibody antigen sites. These and other findings in this study and in the literature suggest that gluten-sensitive enteropathy is immunologically mediated and that IgA antiendomysial antibodies play a significant pathogenetic role.

Anticentromere antibody: an immunological marker of a subset of systemic sclerosis

Our clinical and immunological studies of 114 cases of systemic sclerosis, 54 of Raynauds disease and 46 of other connective tissue diseases, centered on the diagnostic and prognostic significance of anticentromere antibodies (ACA). The ACA occurred in 21 of 84 patients with acrosclerosis, in four of 54 patients with Raynauds disease but in none of 30 patients with diffuse scleroderma or transitional form, acrosclerosis‐diffuse scleroderma, or 46 cases of other connective tissue diseases. The ACA‐positive patients had no contracture or immobilization of the fingers, the indurations and/or indurative oedema were confined to fingers and usually no other types of ANA were detected. However, systemic involvement and the course of the disease were comparable in ACA‐negative and ACA‐positive acrosclerosis patients.

Evaluation of methods for detection of anticentromere antibodies and other antinuclear antibodies

Our immunologic studies of twenty-five patients with acrosclerosis with severe acral involvement, twenty-seven patients with most or all of the signs of CREST syndrome, twenty-two patients with systemic lupus erythematosus as positive controls, and ninety-one blood donors as negative controls centered on an evaluation of eight antigenic substrates, including four types of human cells for the detection of anticentromere antibodies (ACA) and other antinuclear antibodies (ANA). The ACA, which occurred only among the patients with most or all of the signs of CREST syndrome, could be detected reliably on human cell lines HEp-2 and KB but not on a mouse cell line or on the three types of tissue sections examined. Comparisons of human HEp-2 and KB cell lines from four sources indicated that HEp-2 cells are the best of the substrates tested for detection of ACA. Since rodent tissue sections give negative reactions with ACA, they are indicated for confirmation. In general, results varied with the type and source of antigen used. Thus ANA findings need to be expressed not only in terms of the titers of the antibodies and the pattern(s) of their reactions but also in terms of the type of antigen or substrate used, its source, and the diagnostic significance of findings in the given test system.

Specificity of the Crithidia luciliae method for detecting anti-DNA antibodies. Effect of absorption for lipoproteins

Using the immunofluorescent (IF) assay with Crithidia luciliae smears, anti-native (n) DNA antibodies were detected in the sera of 12 of 20 systemic lupus erythematosus (SLE) patients, in 1 of 6 mixed connective tissue disease cases, in 2 of 38 patients with systemic sclerosis but in none of the sera from 96 normal subjects. All anti-nDNA antibodies were associated with antinuclear antibodies (ANA). However, occasionally sera were encountered in routine screening which appear to be positive for anti-DNA antibodies but negative for ANA. Studies of such sera indicate that this is a nonspecific reaction which can be abolished by treating sera with dextran sulfate or heparin. Treatment of SLE sera with these agents had no effect on their anti-nDNA antibody activity. Absorption of sera with Aerosil eliminated the false positive reactions with C. luciliae; however, this treatment also removed immunoglobulins, ANA and anti-nDNA antibodies. Evidence is reviewed which points to a role of complexes of low density lipoprotein and IgG in the nonspecific binding reactions with C. luciliae which is seen as false positive reactions for anti-nDNA antibodies.

Intercellular complement C3 binding in normal skin of patients without pemphigus

We report here direct immunofluorescence studies of normal skin biopsies that exhibited only complement C3 staining in intercellular areas of epithelium with a view toward evaluating the significance of such findings. During a 5-year period, 11,000 skin biopsy specimens were examined for in vivo binding of immunoglobulins, fibrin, and complement C3 by means of defined direct immunofluorescence methods. Four of ten patients demonstrating intercellular C3 deposits alone had drug-related reactions or erythema multiforme, and four were subsequently shown to have a connective tissue disease. Pemphigus was ruled out in all ten cases in these retrospective studies. These observations indicate that the finding of intercellular C3 deposits in the absence of IgG in normal skin is not a sign of pemphigus but, rather, a sign either of an unusual type of drug reaction or, possibly, of some connective tissue disease. However, the finding of intercellular C3 plus IgG (with or without other immunoglobulins) is a sign of pemphigus. The mechanism of C3 deposition without immunoglobulins is not clear.

Effect of erythrocytes treated with enterobacterial common antigen on experimental Salmonella typhimurium infection of mice.

The immunogenicity of enterobacterial common antigen (CA)-treated horse or mouse erythrocytes was determined in Swiss white albino mice by comparing survival rates with control mice, immunized withP. aeruginosa fraction-treated RBC and challenged in parallel with 10 LD50S. typhimurium. The administration of small amounts of CA on horse, but not mouse, RBC significantly delayed mortality; protection was only marginally less than that evoked with 12-fold larger amounts of CA in the absence of RBC. Survival of infected animals was transient; independent of immunogen or control preparation employed, all mice were dead by day 15 after challenge.

Immunologic mimicry between mouse tissue and enterobacterial common antigen.

Organs of Swiss white albino and C57BL/6Ha mice were assessed for an antigen (CRA) which cross-reacts with common enterobacterial antigen (CA). To this end, supernatant fluids (HKS) and ethanol-soluble fractions (ES) of heated homogenates of spleens, kidneys, and livers were examined for their capacities to react with CA hemagglutinins and to engender humoral and cellular events in the rabbit. The immunogenicity of CRA in the rabbit can not be predicted on the basis of CA hemagglutinin neutralization studies alone; although CRA was identified in the liver extracts of both mouse strains, according to this parameter, only the liver fraction of Swiss white albino mice elicited significant numbers of rosette-forming cells (RFC) in the spleens of rabbits. Also, kidney fractions, which primed the rabbits for booster with CA, were less effective in stimulating RFC in the spleens failed to inhibit CA hemagglutination and did not prime rabbits for a CA hemagglutinin response, these same preparations clearly evoked RFC in rabbit spleens. Thus, the antigenicity and immunogenicity of CRA in target organs of mice reflect the mouse strain, extraction procedure, and testing method employed.

Immunologic cross-reaction between enterobacterial common antigen and rat tissue.

Livers, sera, and erythrocytes of MAXX, BN, and Wistar rats were examined for an antigen (CRA) that cross-reacts with enterobacterial common antigen (ECA). Extracts of these tissues were tested for their capacity to reduce anti-ECA titers. By this parameter, CRA was present in certain, although not all, extracts of rat tissue. None of the extracts modified red blood cells (RBC) for agglutination by anti-ECA antisera, nor did MAXX liver extracts engender anti-ECA activity in rabbits. It was speculated that a repressor is present in rat tissue that affects both antigenicity and immunogenicity of CRA. To test this concept, extracts of mixtures of enterobacterial suspensions and liver homogenates were prepared. ECA in the presence of tissue did not modify RBC for anti-ECA activity as did ECA alone, nor did enterobacteria-liver extracts significantly reduce anti-ECA titers. Administration of ethanol-soluble fractions of enterobacteria-liver mixtures to rabbits did not elicit an anti-ECA antibody response as did ECA alone. The finding of CRA in certain rat tissues, sera, and erythrocytes may account, in part, for rats being refractory to immunization with ECA. It is proposed that repressors may mask or abrogate the expression of CRA in animal tissue.