Detection of Alternaria allergens by crossed radioimmunoelectrophoresis

Allergens involved in sensitivity to Alternaria have not been well defined. We used crossed radioimmunoelectrophoresis (CRIE) to study antigens in a crude A. alternata extract, which were capable of binding IgE in human serum. Sera from 35 patients sensitive to Alternaria, 10 not sensitive to Alternaria, and five normal controls were examined. CRIE with hyperimmune rabbit sera demonstrated 22 antigens in the Alternaria extract. After exposure of CRIE gels sequentially with patient serum and 125I-labeled anti-human IgE, autoradiography demonstrated that three of the antigens bound IgE in sera of Alternaria-sensitive subjects. Our results suggest that multiple allergens are involved in A. alternata sensitivity.     

Detection of Alternaria allergens by Western blotting

The technique of Western blotting (immunoblotting) is being increasingly recognized as a means of identifying and characterizing allergens in complex mixtures. Because the allergens are immobilized on nitrocellulose paper, they can be probed for binding of IgE from patients' sera. In addition, the blotting technique provides useful biochemical information, such as the molecular weight of the allergen being studied. In this study we used sodium dodecyl sulfate-polyacrylamide gel electrophoresis to separate antigens in a crude Alternaria preparation, electrophoretically transferred these antigens to nitrocellulose, and incubated the nitrocellulose strips with individual serum and pooled sera from 13 Alternaria-sensitive subjects, sera from three nonallergic individuals, and sera from three individuals allergic to ragweed but not Alternaria. The strips were then washed, incubated with 125I-labeled antihuman IgE, washed again, and exposed to autoradiography film. We observed IgE binding to 20 allergens. Five or more sera from Alternaria-sensitive patients recognized allergens in four molecular weight regions: 55,000 to 60,000, 40,000 to 45,000, 35,000 to 40,000, and 15,000 to 20,000. With pooled sera, the most intense IgE binding was directed at allergens with molecular weights above 40,000. No IgE binding was observed with sera from individuals who were not sensitive to Alternaria. We conclude that Western blotting is an effective method for detecting allergens in complex fungal mixtures.     

HLA antigen frequencies and natural history in Alternaria-sensitive and perennial nonallergic asthmatics

The frequency of HLA-A, -B, and -C loci antigens in random populations of Alternaria-sensitive (N = 100) and perennial nonallergic asthmatics (N = 87) were compared with age- (±5 yr) and sex-matched controls from the same geographic region. There was no association between HLA antigens as measured by frequency analyses and Alternaria-sensitive or perennial nonallergic asthma. Moreover there was no association between HLA antigens and the age of onset of asthma, associated allergic disorders, various environmental factors provoking asthma, total serum IgE levels, and Alternaria-specific IgE antibody.     

Comparison of antigenic and allergenic composition of two partially puriflied extracts from Dermatophagoïdes farinae and Dermatophagoïdes pteronyssinus mite cultures

Dermatophagoïdes farinae (Df 80d) and Dermatophagoïdes pteronyssinus (Dp 80d) extracts were analyzed,for their antigenic and allergenic composition by means of crossed immunoelectrophoresis (CIE) and crossed radioimmunoelectrophoresis (CRIE). By CIE, II antigens could be numbered in Df 80d (Df I to Df II) and seven antigens in Dp 80d (Dp I to Dp 7). This technique allowed us also to define antigens with common as well as specific parts for the two mite species. Among the antigens of D. farinae and D. pteronyssinus, only the antigen corresponding to DO and Dp 5 seems to bear common epitopes to the two mite species, whereas Df 6 and Dp 4 appear to, bear, respectively, specific epitopes 4f each species. Moreover, Df lI appears to bear specific epitopes of D. farinae, although it shows a partial identity with Dp 7. By CRIE, on 20 mite-sensitive patients' sera, we identified, for each mite extract, the allergens responsive to human specific IgE. The allergograms show that the majority of mite-sensitive patients react with Df 11 and Df 6 and with Dp 7 and Dp 4. Tines, these antigens can be considered as major allergens. The minor allergens were also identified. None of these antigens was recognized by the control sera. Moreover, we observed that, for one antigen (antigen 5) there exist antigenic determinants common to the two species of mites toward the rabbit serum and specific allergenic determinants to the human IgE response. A significant correlation was found between the specific IgE binding in CRIE and in RAST (Spearman coefficient: “rs” = 0.61 p < 0.0l ,for Df; “rs” = 0.78 p < 0.01 for Dp).     

Multiplicity of allergens in peanuts

Crude peanut protein fractions from raw and roasted peanuts were examined in the RAST with 10 sera from patients showing clinical peanut sensitivity. The radioactive uptake results, which were generally high, did not reveal any distinguishable pattern. Two commercially available peanut proteins, peanut lectin and phospholipase D, gave poor RAST responses. Three purified peanut proteins, α-arachin, conarachin I, and concanavalin A-reactive glycoprotein, all gave significant RAST results that were generally lower than those obtained with the crude extracts. The extent of RAST inhibition obtained with these materials was inversely related to their abundance in the total peanut protein. Crossed immunoelectrophoresis with extracts from raw and roasted peanut indicated the presence of 22 and 10 anodically migrating antigens, respectively. Sixteen IgE binding antigens were revealed for raw peanut and seven for roasted peanut after incubation with a mixed serum from the 10 patients in crossed radioimmunoelectrophoresis (CRIE) using ¹²⁵I-labeled anti-IgE. CRIE plates treated with individual serum samples showed that all the patients had specific IgE for the major antigen peak, which has been tentatively identified as α-arachin. This major storage protein of peanut, which is known to be particularly heat resistant, may be of greater clinical significance than its apparently low RAST activity would seem to indicate.     

Selected recombinant Aspergillus fumigatus allergens bind specifically to IgE in ABPA

BACKGROUND: Allergic bronchopulmonary aspergillosis (ABPA) is a hypersensitivity lung disease resulting from exposure to Aspergillus fumigatus allergens. Patients with ABPA show elevated Aspergillus-specific serum IgE, a major criterion used in the diagnosis of the disease. Crude culture filtrate and mycelial antigens have been used widely to demonstrate IgE antibody to Aspergillus in the sera of patients. While these antigens have been useful in the diagnosis of ABPA, occasionally they present inconsistency in their reactivity and lack of specificity. Although in recent years, a number of purified A. fumigatus allergens have been produced by molecular cloning, no attempt was made to evaluate them systematically. OBJECTIVE: To evaluate the recombinant proteins from A. fumigatus for their IgE antibody binding, we studied sera from ABPA patients and controls by antigen specific enzyme linked immunosorbent assay (ELISA). METHODS: Recombinant Aspergillus allergens Asp f 1, f 2, f 3, f 4, and f 6 were studied for their specific binding to IgE in the sera of ABPA patients, A. fumigatus skin prick test positive asthmatics, and normal controls from the USA and Switzerland. The sera were blinded and studied by ELISA in two different laboratories. RESULTS: All the recombinant allergens showed IgE antibody binding with sera from patients with ABPA, whereas only fewer asthmatics and normal sera showed significant binding. The three selected recombinant allergens together reacted with all the ABPA patients studied. CONCLUSIONS: The results demonstrate that Asp f 2, f 4, and f 6 can be used in the serodiagnosis of ABPA, while IgE antibody binding to Asp f 1 and f 3 was not specific.     

Quantitative Immunoelectrophoretic Analysis of Extract from Cow Hair and Dander

Quantitative immunoelectrophoresis used for the analysis of a diabased, centrifuged and freeze-dried extract from cow hair and dander revealed 17 antigens. Five of these were identified as serum proteins. Partial identity to antigens of serum and extract from hair and dander of goat, sheep, swine, horse, dog, cat and guinea pig, and to antigens of house dust was demonstrated. Sera from 36 patients with manifest allergy to cow hair and dander selected on the basis of case history, RAST, skin and provocation test, were examined in crossed radioimmunoelectrophoresis (CRIE); sera from five persons with high serum IgE, but without allergy to cow hair and dander, and sera from five normal individuals were controls 31/36 of the sera contained IgE with specific affinity for two of the antigens of the extract. Further, two major and six minor allergens were identified. The control sera showed no specific IgE binding. A significant positive correlation was found between RAST and CRIE for the first group of patients. The approximate molecular weights of the four major allergens obtained by means of gel chromarography were: 2.4 × 10⁴, 2 × 10⁴, 2 × 10³ dalton, respectively. Using Con-A and Con-A Sepharose in crossed immunoaffinoelectrophoresis, eight of the antigens were revealed to contain groups with affinity for Con-A.     

The allergens of dog. I. Identification using crossed radio-immunoelectrophoresis

The antigens present in an extract of dog hair and dander were examined by crossed immunoelectrophoresis (CIE) and the IgE-binding allergens by crossed radio-immunoelectrophoresis (CRIE), respectively, using sera from 60 British and Finnish animal-allergic subjects. The extract was comprised of a minimum of 28 antigens, 11 of which were common to dog serum. IgE antibody in the sera of the dog-sensitive patients bound to 21 of the 28 antigens at varying frequencies and intensities. Binding of any intensity occurred most frequently to two serum proteins: antigen 23 (IgG) binding IgE in 88% of cases, and antigen 3 (dog serum albumin, DSA) in 77% of cases. Dander antigen 8 bound in 63% and antigen 1 in 42% of the sera. Strong IgE binding, however, was most commonly associated with dander antigen 8 followed by antigens 1 and 23 (IgG) then 3 (DSA). The ranking of the antigens as allergens was similar for the two populations except that DSA was more important for the British than for the Finnish subjects.     

Olive (Olea europea) pollen allergens--I. Immunochemical characterization by immunoblotting, CRIE and immunodetection by a monoclonal antibody

The pattern of reactivity of the Olea europea crude extract antigens was analysed after electroblotting to nitrocellulose from SDS-PAGE. The antigens contained in the 17, 19 and 42 K bands were most reactive with specific IgE from individual sera. Following immunization with a crude extract, one monoclonal antibody (OL-1) was raised against components which exhibited IgE binding capacity in electroblotting and crossed radioimmunoelectrophoresis (CRIE). Monoclonal antibody OL-1 reacted with the 17 and 19 K antigens and with three arcs of crossed immunoelectrophoresis (CIE), one of which is considered to contain a major allergen by CRIE.     

Antigens of alternaria. I. Isolation and partial characterization of a basic peptide allergen

A basic peptide allergen has been isolated from Alternaria extracts. Flat-bed gel preparative isoelectric focusing followed by dialysis and lyophilization allowed for concentration of substantial quantities of this heretofore unrecognized allergen. Rocket immunoelectrophoresis and autoradiography (using a 2 X concentrated, IgE serum fraction from Alternaria patient sera and 125I-anti human IgE), a RAST inhibition assay, and skin testing of Alternaria-sensitive patients were used to demonstrate the allergenicity of this fraction. The allergenicity of this protein was sufficient to give 50% RAST inhibition at a concentration of 100 micrograms/ml using standard RAST procedures. The immunoelectrophoresis autoradiography also indicated allergenicity. Skin tests indicated that over 90% of patients hypersensitive to Alternaria crude extracts were also hypersensitive to the basic allergen fraction.     

Identification of allergens in extract of horse hair and dandruff by means of crossed radioimmunoelectrophoresis.

Sera from 26 patients and 4 normals were examined for specific IgE binding to antigens of extract of horse hair and dandruff by means of CRIE. 22 of the patients were RAST- and intracutaneous-positive to horse extract. 4 more of the patients were RAST-negative to horse allergens, but showed allergies to extract of allergens from sources other than horse. The remaining four sera from controls were RAST-negative to horse and had no history of allergy. Antigens of horse hair and dandruff showed a significantly higher degree of binding to specific IgE in the sera from the first group of patients than was the case for the two other groups. A linear correlation between specific IgE binding in RAST and in CRIE was found for the first group of patients. On the basis of these results the major allergens of the examined extract of horse hair and dandruff were identified.     

Studies on IgG subclass antibodies in adult asthma. 1. Serum antigen specific IgG subclass antibodies in asthmatics with late asthmatic response

It is clear that immediate asthmatic response is mediated by IgE-dependent mechanisms. However, late asthmatic response is induced by inhalation of antigens without antigen specific IgE antibodies in some asthmatics, especially in intractable asthma induced by Candida antigen. To elucidate the relationship between those bronchial responses and antibodies, antigen specific IgG subclass antibodies in sera from asthmatics were measured and compared with IgE antibody. The results were as follows. 1. Avidin-biotin ELISA (enzyme-linked immunosorbent assay) method was established for the measurement of specific IgG and IgG subclass antibodies to mite or Candida antigen. 2. Serum levels of antigen specific IgG and IgG1 antibodies in asthmatics with LAR provoked by mite or Candida antigen were significantly higher than those in asthmatics without LAR (p less than 0.01). 3. Serum levels of specific IgE antibody to these antigens in asthmatics with LAR provoked by mite or Candida antigen were slightly lower than those in asthmatics without LAR, though the difference is not significant. These results suggest that high serum levels of specific IgG and IgG1 antibodies to these antigens play a role in inducing LAR in asthmatics with LAR.     

Investigations of culture medium-free house dust mites. III. Antigens and allergens of body and fecal extract of Dermatophagoides farinae

Crossed immunoelectrophoresis (CIE) and crossed radioimmunoelectrophoresis (CRIE) were used to characterize antigens (Ags) and allergens derived from Dermatophagoides farinae (DF) culture media-free mite body and mite fecal matter extracts. CIE of DF body and DF feces extracts revealed the presence of 35 and 20 Ags, respectively. CRIE experiments demonstrated IgE binding by 14 and seven DF body and DF feces Ags, respectively, when CIE gels were incubated with reference sera from clinically mite-sensitive patients. Binding of specific IgE to the various Ags in the two extracts varied significantly both in frequency and in strength from patient to patient and within the same patient's serum. Sera from some patients demonstrated IgE binding predilection for specific DF body Ags, whereas other sera exhibited greater binding preference for DF feces Ags. Homologous, heterologous, and intermediate gel CIE and CRIE clearly demonstrated that DF bodies and DF feces share some common Ags or epitopes, but the two different extracts also were quantitatively different. Some Ags and allergens originate from mite body material and are not present in mite feces. These results indicate that only extracts containing high concentrations of both body and fecal allergens should be used in clinical testing and therapy.     

Alternaria spore and mycelium sensitivity in allergic patients: in vivo and in vitro studies.

Extracts from Alternaria spores and mycelia were prepared to evaluate their allergenic potencies in allergic patients. Twelve Alternaria-sensitive patients, with histories of rhinitis or asthma, were submitted to skin prick tests and five of 12 patients received nasal challenges with spores and mycelia. In vitro, the allergenic activity of each extract was determined by RAST, basophil histamine-release and RAST-inhibition. All patients demonstrated skin reactivity to both extracts while skin reactivity to mycelia was higher than that induced by the spore extract (P < .005). Of the 12 patients, 11 had positive mycelia-RAST and 9/12 had positive spore-RAST. It was found that mycelium-IgE antibody levels were higher than spore-IgE antibody levels (P < .005). Nine RAST-positive patients had positive histamine release tests (> 50%) and basophils challenged with mycelia appeared 10-fold more sensitive compared to the spore challenge. In four of five patients subjected to nasal provocation tests, immediate-type rhinitis was elicited either after spore or mycelium challenge. The patients exhibited a higher nasal reactivity with the mycelium challenge than with spores. RAST-inhibition studies demonstrated that mycelial extracts shared common allergens with spore. These results indicated that Alternaria spore and mycelium were potent allergens in allergic patients and there was a variability in the pattern of in vivo and in vitro reactions between the patients for each allergen.     

Immunotherapy in Hay Fever with Two Major Allergens 19, 25 and Partially Purified Extract of Timothy Grass Pollen

Grass pollen hay fever patients were hyposensitized in a prospective 3-year double blind study with two timothy extracts. Group WPA (20 patients) was treated with partially purified extract = whole pollen allergens (WPA) (Alutard®SQ) and group PPA (20 patients) was treated with a purified pollen allergen (PPA) = two isolated major allergens, Ag19 and Ag25. Both aluminiumhydroxide adsorbed extracts were biologically standardized. Clinical results from the first season have been recently published and WPA showed significantly fewer symptoms (P= 0.001) than PPA. Corresponding preseasonal and seasonal in vitro results are presented here. Serum total IgE, specific IgE versus total and individual allergens of timothy pollen and allergen-specific IgG showed a rapid increase in both groups until the season but showed no further increase or decrease during the season. Specific IgE was shown to correlate to specific IgG during hyposensitization. Group WPA, with fewer symptoms in grass pollen season than group PPA showed a significantly higher increase of specific IgG and IgE than group PPA, but individual symptom scores were not correlated to specific IgG or IgE, or their ratio. Specific IgE and IgG increases were not correlated to dosage. Surprisingly, almost all females were low-responders to specific IgG and IgE though they had equal symptom scores, dosage, and side effects as males, while the characteristics of high-responders to antibody were: youngest individuals and shortest duration of symptoms prior to treatment. Crossed radioimmunoelectrophoresis (CRIE) showed specific reaction to stimuli but no development of allergy against new timothy antigens. High response of IgE to Ag 19 in CRIE during initial hyposensitization seems suitable as marker for prospective evaluation of clinical effect in grass pollen hyposensitization. Nasal secretion was collected after methacholine provocation, and total IgE and specific IgE detected. There was no response to treatment, only a slight increase during the season. No decrease in nasal reactivity to methacholine was noted during one preseasonal hyposensitization.     

IgE and IgG antibody responses to recombinant Alt a 1 as a marker of sensitization to Alternaria in asthma and atopic dermatitis

BACKGROUND: Sensitization to Alternaria alternata is a risk factor for the development of wheezing and asthma. Alt a 1 is the major Alternaria allergen causing sensitization in asthmatics. Some atopic dermatitis (AD) patients have very high immunoglobulin (Ig)E antibody (ab) to Alternaria as analysed by Pharmacia CAP, however, it is not clear whether these are specific responses or whether Alt a 1 is involved in disease symptoms. OBJECTIVE: The aim of this study was to analyse specific IgE and IgG ab responses to recombinant Alt a 1 in asthmatic and AD patients and to compare these results to IgE ab against Alternaria measured by CAP. METHODS: Sera from individuals who were IgE positive to Alternaria by CAP were obtained from 58 patients with asthma/rhinitis, 19 patients with AD, and 20 patients with cystic fibrosis (CF) who were included as specificity controls. IgE and IgG ab to recombinant Alt a 1 were measured by radioimmunoassay (RIA). RESULTS: Of 43 asthma/rhinitis patients having an Alternaria CAP score > 2, a high percentage (93%) had both IgE and IgG ab to Alt a 1, emphasizing its importance as a major allergen. Only, 47% of AD patients with CAP score greater than 2 had ab to Alt a 1, and their levels were low when compared to the asthmatics. For CF controls, 75% of these patients had no IgE ab to Alt a 1, and those which were positive to Alt a 1 by RIA were also positive by CAP. Overall, patients with a low CAP (1-2) had a low prevalence (20-30%) of IgE or IgG ab to Alt a 1. CONCLUSION: IgE and IgG ab to Alt a 1 in asthmatics are good markers for sensitization to Alternaria. Although AD patients gave high Alternaria CAP scores, they had low or undetectable levels of IgE to Alt a 1, suggesting that other Alternaria allergens may be important in AD or that the CAP results are non-specific. Recombinant allergens may provide more specific measures of sensitization to fungi.     

Immunological features of asthma (Part II)

A serological comparison was made of two groups of 120 matched asthmatic and healthy subjects, between the ages of 20 and 49 years and matched for age and sex, in terms of serum total levels of IgG, IgM, IgA. IgD and IgE and of specific antibody levels in each immunoglobulin class to five common UK allergens. The relationship of clinical features to the serological tests was also examined in the asthmatic subjects. The following statistically significant findings were shown. The patients had higher levels than the controls of total globulins and of IgG, IgA and IgD but not IgM. In both patients and controls the females had higher IgM levels than the males. The total IgE levels were higher in patients than in the controls and the male patients had higher levels than the female patients. Total IgE levels were also related, to the numbers of first degree relatives with asthma, hay fever and eczema, to the severity of hay fever and to the amount of time off work in the male patients. In those male patients with exercise induced asthma the total IgE levels were lower than in those not showing this reaction. As for the other iminunoglobulins, the only significant differences were a higher [gG level in patients with FEV, or PFR>50% predicted and a higher IgD level in patients with hay fever. Radio-immunodiffusion tests for specific precipitins were positive for Dermatophagoldes pteronyssinus in comparable numbers of asthmatics (25.8%) and controls (21.7%). Positive precipitin tests were uncommon in tests with extracts of grass pollen, Aspergillus fumigatus, cat and dog hair in the patients and even less so in the controls. Positive RAST tests for specific IgE antibodies were obtained in patients and controls respectively, against D. pteronyssinus 59% and 11% grass pollen 37.0 and 12%, and A. fumigatus 6% and 4%. The male patients showed the closest significant relationship of specific IgE to D pteronyssinus and the history of house dust allergy, positive skin test and nasal test. in the females only the skin and specific IgE tests were related. Both sexes showed a significant association between specific IgE to grass pollen and positive skin tests and nasal tests, but only the males showed an association with the history. The size of skin test weal to D. pteronyssinus were related to the levels of specific IgE antibody, Correspondence: Professor J. Pepys, Cardiothoracic Institute, Brompton, London SW3 6HP. No differences were found between the four skin test groups and between the asthmatics and the control subjects in the incidence of bacterial precipitins and auin-antibodies.     

Allergic bronchopulmonary aspergillosis: Reactivity of IgE and IgG antibodies with antigenic components of Aspergillus fumigatus (IgE/IgG antigen complexes)

Sera of patients with ABPA were tested by XRIE tests incorporating their own serum (self-XRIE) to detect the presence of IgG/IgE antigen complexes to a “reference” Aspergillus fumigatus preparation. Of the 32 sera studied, 29 (90%) had visible precipitin (IgG) peaks, and 27 of these 29 as well as the three apparently precipitin-negative sera, i.e., 30 (94%), showed binding of specific IgE by autoradiography. The two precipitin-positive sera that did not show IgE binding were also skin test negative and RAST negative to this A. fumigatus antigen. Specific IgG as determined in ELISA correlated well with the grading of the XIE precipitin peaks (p < 0.05). There was also a highly significant correlation between specific IgE by RAST and grading the radioactive uptake seen in the autoradiograph (p < 0.001) indicating, for each serum, the presence of IgG antibodies to most of the components to which there was specific IgE. In the self-XRIE tests there was considerable variation of reactivity from serum to serum, in numbers of antigen/antibody peaks observed, in relative peak heights, and in the intensity of the respective staining. By comparing each test to a “reference” pattern developed with the use of an ABPA serum pool, the antigenic components of A. fumigatus were found to be of two main types: (1) antigens that appeared to be poorly precipitating (possibly low-molecular-weight components) but showed strong IgE binding (these were apparently major allergenic components and with one exception proved to be the faster migrating components) and (2) antigens that produced the strongest precipitin reactions with only weak binding of specific IgE and therefore minor allergenic components.     

Immunochemical partial identity between two independently identified and isolated major allergens from Alternaria alternata (Alt-I and Ag 1)

Partially purified preparations of Alt-I, the main allergenic fraction of Alternaria alternata isolated by Yunginger, and of Ag 1, shown in crossed radioimmunoelectrophoresis (CRIE) to be the dominating major allergen of A. alternata (Løwenstein, Nyholm), were compared by tandem crossed immunoelectrophoresis (CIE), RAST inhibition, and the CRIE-related technique, single radial radioimmunodiffusion (SRRID). The two allergen preparations showed reaction of identity in tandem-CIE and indistinguishable specific IgE binding in CRIE and SRRID, regardless of antibodies and serum pools used. In RAST inhibition, the relative potencies of the allergen preparations and of the crude extracts correlated well with their Alt-I/Ag 1 content as estimated by rocket immunoelectrophoresis. Moreover, all inhibition curves were parallel, confirming identical IgE binding by Alt-I and Ag 1 with the serum pools used. A second preparation of Alt-I, isolated from another strain of Alternaria, showed reaction of partial identity with Ag 1 in tandem-CIE, indicating that different variants of Alt-I (Ag 1) may exist in different strains of A. alternata.     

HLA-A, B and C and HLA-DR antigens in intrinsic and allergic asthma

Some 103 patients with asthma and 100 healthy volunteers have been typed for HLA-A, B and C and HLA-DR antigens. The 103 patients consisted of thirty-three with intrinsic asthma, thirty-four with extrinsic asthma, and thirty-six known to have precipitins to Aspergillus fumigatus. No increase in frequency of any of the A, B, C, or DR antigens was found to be significant after correction for the number of comparisons was made. However certain trends comparable to findings in other immunopathic disorders were noted. For example B12 was increased in the allergic asthmatics (46 vs 29% controls) and it is suggested that B12 is associated with the ability to produce the IgE antibodies. A3/B7/DRw2 (which are in linkage disequilibrium) all show a decreased frequency in intrinsic asthma (24, 12 and 9%vs 32, 26 and 24% respectively in controls). Finally B8 and DRw3, which showed a moderate increase in frequency in all three groups of asthmatics, were found in five of seven patients with low atopy but persisting antibodies to A. fumigatus. Further detailed studies of these asthmatic subgroups is warranted.