Anti-IgA in Selective IgA Deficiency

Anti-IgA antibodies were found in 14 of 33 (42%) IgA-deficient donors. In healthy IgA-deficient blood donors anti-IgA appeared associated with the presence of HLA DR3. The antibodies were mainly of the IgG1 and, in high-titred sera, IgG4 subclasses. Sera containing high-titred anti-IgA selectively impaired IgA synthesis in vitro as induced by direct and indirect polyclonal B-cell activators. These antibodies may play a role in the pathogenesis and/or the maintenance of IgA deficiency.     

Anti-IgA antibodies in pregnancy

A survey of 28,000 pregnant women revealed an incidence of IgA deficiency (serum IgA less than 1 mg per deciliter) of 1 in 450, which is identical to that in a normal blood-donor population of both sexes. Using an enzyme-linked immunosorbent assay (ELISA) in a study of 61 serum samples from IgA-deficient pregnant women, we observed antibodies to IgA2 alone in 20 per cent, as compared with 7.5 per cent of pregnant women not deficient in IgA and no IgA-deficient blood donors. Antibodies reacting with IgA1 alone were present in occasional serum samples (2 to 7 per cent) from all groups studied, and class-specific anti-IgA antibodies were present in 17 per cent of IgA-deficient blood donors and in 16 per cent of IgA-deficient pregnant women. Blocking experiments showed that some serum samples contained an antibody that reacted with both IgA1 and IgA2, whereas others contained two antibodies, one reacting with IgA1 and the other with IgA2. The anti-IgA2 antibodies tended to diminish in titer after delivery. The ELISA was, as expected, more sensitive than the hemagglutination assay. The offspring of IgA-deficient mothers (but not of IgA-deficient fathers) had levels of serum IgA below the normal mean (21 of 27); 12 had levels more than 1 S.D., and seven had levels more than 2 S.D., below the normal mean. Of the seven infants with serum IgA levels more than 2 S.D. below the normal age-related mean, five had mothers with anti-IgA antibodies during gestation. It is possible that maternal anti-IgA exerts a transplacental effect on the fetal immune system, causing IgA deficiency in some instances.     

HLA and IgA deficiency in blood donors

We have identified 67 IgA deficient healthy blood donors in our region by systematic screening of 24,782 blood samples. HLA typing results on 36 of these donors indicated a significant association with both HLA-A1 and HLA-B14 antigens. The frequency of A29 and B8 antigens was also increased. However, B8 association may be secondarily involved due to linkage disequilibrium (e.g., A1-B8-DR3). The frequency of HLA-A1 and HLA-B8 antigens was increased in the group of IgA deficient donors who developed anti-IgA antibodies (53.9%) compared to those who did not (26.1%). Although the sample size appears to be too small to show a statistical significance (chi 2 = 2.76, 0.05 less than p less than 0.1), this is the first report to imply a possible HLA association with anti-IgA antibody formation by IgA deficient healthy individuals.     

Recurrent extended HLA haplotypes in children with selective IgA deficiency

HLA supratypes as well as serum IgG, IgA and IgM levels were determined in 44 children and adolescents with severe IgA deficiency (serum IgA less than 5 mg/dl) and in first degree relatives. Frequencies of the HLA alleles B14, DR1, DQW1, C4A2 and C4B2 were significantly higher in the IgA-deficient patients than in the controls. The most recurrent haplotype among patients was B14, DR1 (p less than 10(-4) preferentially associated with A33, A28 or A blank. The supratype B14, Bfs, C4A2, C4B2, DR1, DQW1 was present in a 14-fold higher frequency than in the controls, and strongly suggests the presence of a gene in the HLA region involved in the deficiency of IgA. The fact that this supratype was not always associated with IgA deficiency in the parents and that not all IgA-deficient subjects had this supratype is discussed. Severe IgA deficiency was found in four mothers (two of the four mother-child pairs shared the B14, DR1 haplotype); three other relatives (one father and two brothers) had partial IgA deficiency and did not have the B14, DR1, DQW1 haplotype.     

Anti-IgA antibodies in IgA-deficient children

IgG and IgM isotype antibodies to polyclonal human IgA, myeloma IgA₁, and myeloma IgA₂ were estimated in 38 IgA-deficient children aged between 0.9 and 15 years. All children had IgM anti-IgA antibodies. IgG antibodies against either polyclonal IgA, IgA₁, or IgA₂ were present in 63% of the IgA-deficient children. IgG anti-IgA antibodies were detected against all three antigens in 8 of 11 severely IgA-deficient children and in 7 of 27 partially IgA-deficient children, but in only 1 of 23 healthy adult controls. The proportion of children with IgG anti-IgA antibodies was significantly greater in the severely IgA-deficient group in comparison with the partially IgA-deficient group and the adult controls (chi-square test,P<0.01 andP<0.005, respectively). There was a strong correlation within each IgG subclass between antibody responses toward each of the three IgA antigens. Twenty-four children were followed over a period ranging from 0.9 to 11 years (mean, 2.3 years). Three children who were initially IgG anti-IgA antibody negative became antibody positive and three who were antibody positive became antibody negative. Five children with severe IgA deficiency remained severely IgA deficient and IgG antibodies to IgA persisted in all five at follow-up. The presence of IgG anti-IgA antibodies did not influence the normalization of serum IgA at follow-up in 14 of 19 children who were initially partially IgA deficient.     

HLA association of anti-IgA antibody production

We have HLA typed 46 unrelated IgA deficient blood donors in our region. A weak association to DR3 (RR = 2.07) was observed. However, owing to the heterogeneous nature of this group, we found DR7 may be associated with those who developed class-specific anti-IgA antibodies (RR = 2.94), whereas DR1 may be associated with those who did not (RR = 2.42).     

HLA antigens in IGA deficient paediatric patients

HLA antigens (A, B, C and DR loci) were studied in 62 IgA-deficient (IgAd) paediatric patients: 17 with coeliac disease (CD), 13 with juvenile arthritis (JA), 27 with frequent respiratory tract infections (RTI) and five with other diseases. The frequencies of HLA antigens in IgAd patients were compared with those in healthy blood donors, and in CD and JA patients with normal serum IgA levels. The IgA deficiency in the patients showed significant associations with HLA A1, B8, B13, Cw6, DR3 and DR7 (P less than 0.0005, P corr less than 0.02 vs controls) and decreased frequencies of DR2 (P less than 0.0005, P corr less than 0.02 vs controls). The HLA associations typical of coeliac disease, increased frequencies of HLA-B8 and DR3, were evident among the IgAd coeliacs; in contrast to the coeliacs with normal IgA levels, the IgAd coeliacs showed a significant increase of the HLA-Cw6 allele (P less than 0.0005, P corr less than 0.02 vs control coeliacs). Increased frequencies of HLA-A1, B8, B13, Cw6, DR3 and DR7 were noted in the patients with RTI, which can be explained by the frequent occurrence of the haplotypes A1, B8, DR3 and B13, DR7, the latter haplotype often also having the Cw6 allele. Among the IgAd JA patients, the antigen frequencies were similar to those in the JA patients with normal serum immunoglobulins.     

IgG anti-IgA1 and anti-IgA2 antibodies: their measurement by an enzyme-linked immunosorbent assay and their relationship to disease

IgG anti-IgA antibodies were measured by an enzyme-linked immunosorbent assay using one IgA1 and one IgA2 (m1) myeloma and a pooled IgA protein preparation as antigens. Class-specific anti-IgA antibodies occurred in 0.8% of non-IgA-deficient sera and 24.3% of IgA-deficient sera. Antibodies reacting with IgA1 only occurred in 2.6% of non-IgA-deficient sera and 6.7% of IgA-deficient sera. Antibodies reacting with IgA2 only occurred in 0.6% of non-IgA-deficient sera and 2.7% of IgA-deficient sera. The prevalence of anti-IgA in IgA deficients with inflammatory disease was higher (81.8%) than in IgA deficients without disease (24.1%) and was accounted for by class-specific antibodies.     

C4 polymorphism and major histocompatibility complex haplotypes in IgA deficiency: association with C4A null haplotypes.

IgA-deficient individuals (n = 110) and six families comprising 9 cases of IgA deficiency were typed for HLA-A, -B, -DR, C4 and factor B. Phenotype frequencies were increased for HLA-B8 (p = 0.004), HLA-DR3 (p = 0.001) and homozygous C4AQ0 (p = 0.01) and decreased for HLA-B7 (p = 0.004), HLA-DR2 (p = 0.0001) and C4A3 (p = 0.00007) compared to controls. Homozygous C4A deficiency was found in 20% of IgA-deficient persons. As clearly suggested by investigation of families, the findings could be attributed to high prevalence of the extended major histocompatibility complex (MHC) haplotype [HLA-A1, B8, C4AQ0, C4B1, BfS, DR3] in IgA deficiency. All but 1 of the 9 IgA-deficient persons included in the family study carried this haplotype and 4 of them were homozygous. In the families, 3 persons with normal serum IgA concentrations had the same MHC haplotypes as their IgA-deficient relatives. The findings were also consistent with possible overrepresentation of other MHC haplotypes with aberrant C4 gene organization in IgA deficiency. As previously suggested, the presence of two MHC haplotypes associated with IgA deficiency appears to be a necessary but not sufficient requirement for manifestation of the condition. The putative existence of a recessive gene in the MHC with regulatory function with regard to IgA gene expression is consistent with the findings.     

HLA-A, B, Cw and DRB1, DRB3/4/5, DQB1, DPB1 frequencies in German immunoglobulin A-deficient individuals

HLA class I and II frequencies and haplotype frequencies were determined in 80 German immunoglobulin (Ig)A-deficient individuals and 157 healthy controls with normal IgA levels using serological and DNA typing methods. For several alleles, significant associations were found, which could be explained mainly in the context of a positive association with three different extended haplotypes (HLA-B*08:DRB1*0301: DQB1*0201, HLA-B*14:DRB1*0102:DQB1*0501 and HLA-B*44:DRB1*0701:DQB1*0202) and a negative association with a fourth haplotype (HLA-B*07:DRB1*1501:DQB1*0602). Furthermore, for the first time this study reports a positive association of IgA deficiency with DPB1 alleles. Homozygosity rate for the gene loci DRB1 and DQB1 was increased in IgA deficiency. Further analysis suggested a different pattern of HLA associations depending on the degree of IgA deficiency and the gender of the IgA-deficient individuals.     

Anti-IgA antibodies in selective IgA deficiency and in primary immunodeficient patients treated with gamma-globulin

Sera from 106 blood donors, 40 patients with primary immunodeficiencies (ID) treated with gamma-globulin, and 46 patients with selective IgA deficiency were analyzed by an enzyme-linked immunosorbent assay for anti-IgA antibodies. Increased levels of antibodies to IgA were found in 5.6% of the blood donors, 17.5% of the ID patients, and 36.8% of the isolated IgA deficiencies. The percentage was higher in patients with IgA and IgG2 deficiencies (50%). The percentage of patients having increased levels of anti-IgA antibodies was similar to the total prevalence of the 10 other autoantibodies studied. These anti-IgA antibodies were mainly of the IgG class, except from one blood donor with IgM antibodies, and two patients, one with isolated IgA deficiency and the other with common variable immunodeficiency who had anti-IgA antibodies of the IgE class. The latter patient developed a near fatal anaphylactic reaction when intravenous gamma-globulin was administered. Most of the patients with severe adverse reactions to gamma-globulin did not present anti-IgA antibodies. Our data suggest that at least in some immunodeficient patients the elevated amounts of anti-IgA antibodies are not related to the administration of exogenous IgA. The importance of measuring anti-IgA antibodies of the IgG and IgE isotypes in IgA-deficient patients as well as in patients in treatment with gamma-globulin is emphasized.     

Serum IgG and IgG Subclass Contents in Different Gm Phenotypes

Different serum IgG and IgG subclass levels were found among Gm phenotypes of a normal population. One hundred and fifty-seven Caucasian blood donors were investigated for the reciprocal Gm allotypes on IgG subclass loci namely: for IgG1, Glm(f) and Glm(a); for IgG2, G2m(n) and G2m("); and for IgG3, G3m(b) and G3m(g), and subgrouped in the seven most common Gm phenotypes. The frequencies of Gm phenotypes and haplotypes were given, including numbers of the previously little known G2m(n,") heterozygous individuals. Mean serum quantities ±SD and range of IgG, IgG1, IgG2, IgG3 and IgG4 were given for different Gm phenotypes. The IgG content was significantly lower in the Gm(f,",b/f,",b) phenotype in which the IgG2 levels were also significantly lower, compared with values of the other phenotypes. IgG3 levels were significantly lower in the Gm(a,",g/a,",g) phenotype compared with other phenotypes. These data imply the importance of Gm(f,",b/f,",b) and Gm(a,",g/a,",g) phenotypes causing lower amounts of IgG antibodies. In evaluating IgG subclass deficiency, the range for the low responding Gm(f,",b/f,",b) and Gm(a,",g/a,",g) phenotypes should be considered.     

HLA-A, B, C and DR antigens in immunoglobulin A deficiency

HLA-A, B, C and DR typing was performed in 21 unrelated healthy blood donors with selective IgA deficiency (< 0.02 G/l of IgA). A significant increase in HLA A1 ( P < 0.05), HLA B8 ( P < 0.01) and HLA DR3 ( P < 0.001) was found. The frequency of HLA A28 was also increased ( P < 0.05). Furthermore, HLA A28 was present in all four donors lacking DR3.     

The role of anti-IgA antibodies in causing adverse reactions to gamma globulin infusion in immunodeficient patients: a comprehensive review of the literature

Anaphylactic reactions to immunoglobulin infusions in immunodeficient patients with undetectable IgA have been attributed in several reports to IgG or IgE anti-IgA antibodies. However, other reports have not supported an association between such antibodies and the development of severe reactions. We have reviewed the articles reporting reactions to immunoglobulin products in IgA-deficient patients, as well as those describing the presence of such antibodies in the absence of reactions to infusions. A?variety of factors might influence the association of adverse reactions with anti-IgA antibodies, including the serum concentration and isotype (IgG or IgE) of the anti-IgA antibody, its specificity (class or subclass specific), the method of measurement, and the IgA content of the gamma globulin infusion and its route of administration. The role of anti-IgA antibodies in causing anaphylaxis in IgA-deficient patients receiving gamma globulin therapy is still controversial. Larger (multicenter) studies are needed to further evaluate this association.     

Review: IgA anaphylactic transfusion reactions. Part I. Laboratory diagnosis, incidence, and supply of IgA-deficient products

Despite yielding a definitive diagnosis in fewer than 20 percent of anaphylactic transfusion reactions, investigation for IgA deficiency and the presence of presumably pathogenic IgG anti-IgA is useful in patient management. Individuals with demonstrated anti-IgA are thereafter committed to receiving IgA-depleted cellular products or IgA-deficient plasma and derivatives to prevent recurrent severe reactions. Unfortunately, in populations of IgA-deficient individuals screened for anti-IgA, the predictive value of the test in the absence of a prior reaction is quite low. Anti-IgA testing is complex and limited to a few reference laboratories, many of which still employ a labor-intensive hemagglutination assay developed in the late 1960s. Timely decisions regarding further transfusion management of patients experiencing anaphylaxis often rely upon more rapidly obtained assays of the IgA concentration as an indicator of the likelihood of subsequent demonstration of anti-IgA. The scarcity of IgA-deficient banked plasma products and dedicated plateletpheresis donors has led to the development of American Rare Donor Program policies designed to appropriately allocate these precious resources. The test methods used to establish the diagnosis of IgA deficiency and identify the approximately one third of these individuals with anti-IgA are discussed, along with the incidence of abnormal tests in various populations. Also presented are testing recommendations for the identification of an IgA-mediated mechanism for transfusion-associated anaphylaxis and qualification of patients to receive rare IgA-deficient plasma-containing products.     

Cross-reactive antibodies induced by xenogeneic IgA can cause selective IgA deficiency

Selective immunoglobulin A deficiency (sIgAD) is the most common immunodeficiency in humans. Auto-reactive antibodies to human immunoglobulin A (IgA) are found in the serum of 20–40% of individuals with sIgAD. It is unknown whether these antibodies play a role in the pathogenesis of this immunodeficiency and although the prevailing thought is that they are secondary to the onset of sIgAD, there is very little, if any, support for this notion. Here, we propose that anti-IgA antibodies are in fact responsible for the removal of IgA from serum, and that the inducing antigen is most probably a xenogeneic IgA. This hypothesis is based on data obtained from an sIgAD patient in whom changes in dietary consumption of beef and/or bovine dairy products resulted in changes in anti-IgA levels in the serum. To test the hypothesis, the presence of anti-bovine IgA antibodies was tested by a highly specific enzyme-linked immunosorbent assay in serum samples from IgA-deficient and control individuals. All 13 sIgAD individuals with anti-IgA antibodies had a higher titer against bovine IgA than against human IgA. Of 23 control individuals, a surprisingly high proportion (65%) was also found to have IgG anti-bovine IgA antibodies. These results support the hypothesis that the anti-human IgA antibodies found in IgA-deficient individuals are originally produced against bovine IgA. These antibodies are found in many normal individuals, but only in cases where they cross react with endogenous human IgA, sIgAD may develop.     

Immunoglobulin allotypes in rheumatoid arthritis

The HLA region of chromosome 6 may account for as little as 20% of the genetic contribution to rheumatoid arthritis (RA). Previous studies have shown associations with immunoglobulin heavy chain allotypes (Gm) coded for by genes on chromosome 14. We review evidence that G1m(x)-bearing haplotypes are associated with DR4-positive, but not DR4-negative RA. A differing Gm association has been reported in two studies of RA patients with circulating antibodies to native type II collagen, which might constitute a specific genetic subgroup. A family study has shown no evidence for genetic linkage between Gm allotypes and susceptibility to RA. It is possible that Gm (or linked 14th chromosomal genes) may influence susceptibility to RA by interacting with different genes within the HLA region. Immunoglobulin light chain allotypes (Km) coded by genes on chromosome 2 show no consistent association with RA.     

Interplay of immunoglobulin G heavy chain markers (Gm) and HLA in predisposing to systemic lupus nephritis

We have studied the distribution of IgG heavy chain markers (Gm) among 90 Hungarian patients with systemic lupus erythematosus (SLE) (55 of whom were also typed for HLA). This study confirms previously described increases in HLA-B8 and DR3 in this condition. No difference in the distribution of Gm phenotypes was found between patients and 168 controls from the same geographical area. HLA-B8/Gm homozygous individuals were, however, at greater risk for SLE (relative risk = 5.13) compared to B8 + Gm heterozygotes or B8- individuals, irrespective of Gm phenotype. When patients with renal manifestation (n = 40) were compared to those without, the Gm phenotype 3; 5, 13 was found to be significantly increased (chi 2 = 10.36, P less than 0.0001, relative risk (RR) = 4.69). HLA and Gm increased additively the risk for renal manifestations in that for those patients who were both Gm3;5,13+ and HLA-B8+, PR was 110, while it was 21.2 for Gm3;5, 13-/B8+, 7.9 for Gm3;5, 13+/B8- and 1.0 for Gm3;5, 13-/B8- patients. The study suggests that combined HLA and Gm typing can be used to identify SLE patients at high risk for manifesting renal abnormalities.     

Discordance between IgA Switching at the DNA Level and IgA Expression at the mRNA Level in IgA-Deficient Patients

IgA deficiency is a common immune disorder in Caucasians and is associated with certain MHC conserved extended haplotypes, such as [HLA-B8, SC01, DR3], which presumably carry a susceptibility gene(s). We applied a competitive digestion-circularization PCR method to quantitate the number of switch (S)mu to S alpha rearrangements in peripheral B cells from IgA-deficient subjects homozygous for this haplotype and compared their number with the productive C alpha mRNA level to determine C alpha gene expression in IgA-switched B cells. Two types of defects, low expression of both secreted and membrane forms of productive C alpha mRNA in IgA-switched B cells and impaired IgA switching, were characterized in IgA-deficient subjects homozygous for [HLA-B8, SC01, DR3]. The former defect was also found in another noncarrier subject. It may directly cause low IgA secretion and reflects a blockade in post-IgA switch differentiation of B cells. These results suggest that the heterogeneity of defects in IgA deficiency is not simply ascribable to MHC susceptibility genes.     

Long-term follow-up of anti-IgA antibodies in healthy iga-deficient adults

A follow-up study of anti-IgA antibodies in 159 healthy blood donors with severe deficiency of serum IgA (<0.05 mg/L) and in 45 donors with decreased serum IgA levels (0.05–799 mg/L), identified in 1971–1980, was carried out. Initially anti-IgA antibodies were determined by a hemagglutination (HA) method and two reexaminations were done in 1990–1992 by an enzyme immunoassay. The median follow-up period was 19 years, during which anti-IgA level was changed considerably in only four persons, increased in two, and high level antibodies (>1/1000 by HA) appeared in two. In reexaminations anti-IgA antibodies were found in 30 (19%) subjects with severe IgA deficiency and the antibody levels remained relatively constant in those who had high and medium antibody levels. Anti-IgA antibodies were not found in subjects with decreased, but detectable serum IgA. Thus it seems that only those healthy adults who have severe IgA deficiency develop anti-IgA antibodies and their anti-IgA levels remain fairly constant Of the 159 subjects with severe IgA deficiency, 66 had a history of IgA exposure, but no correlation to anti-IgA development was noted.Portions of the work have been presented in a poster form at the XXIIIrd Congress of the International Society of Blood Transfusion in Amsterdam, the Netherlands, July 2–9, 1994.