Scand. J. Immunol. 36, 279-283, 1992
Aberrations in Titre and Avidity of Serum IgM and IgG Antibodies to Microbial and Food Antigens in IgA Deficiency F. CARDINALE*§, V. FRIMANf. B. CARLSSON*, J. BJORKANDERJ L. ARMENIO§ & L. A. HANSON* Departments of *ClinicaI Immutiology, tinfectious Diseases and Jlnternal Medicine I, The Asthma and Allergy Research Center. University of Goteborg, Goteborg, Sweden, and §Department of Clinical and Social Paediatrics, University of Bari, Bari, Italy
Cardinale F, Friman V. Carlsson B, Bjorkander J, Armenio L. Hanson L A . Aberrations in Titre and Avidity of Serum IgM and IgG Aniibodies to Microbial and Food Antigens in IgA Deficiency. Scand J (mmunol l992;36:279-83 The antibody levels and relative avidity of serum IgM and IgG aniibodies against E. coli O antigens, poliovirus type 1 and /J-lactoglobulin were determined with enzyme-linked immunosorbent techniques in IgA deficient (IgAd) patients with frequent respiratory tract infections and healthy IgAd individuals. Heallhy individuals with normal tmmunoglobulin levels served as controls. The IgM antibody levels against the bacterial, viral and food antigens and the IgG antibody levels against the bacterial antigens were significantly higher in the IgAd group with recurrent infections than in the group of healthy IgAd individuals. The symptomatic IgAd group had significantly higher levels of the IgG antibodies against the bacterial antigen, also when eompared with controls. In contrast the healthy IgAd individuals had the highesi avidities of IgM antibodies to the viral and food antigens. The high avidities of aniibodies could be a compensatory host defence mechanism in IgAd. These aberrations may appear as a consequence of increased mucosal exposure in IgAd to antigens such as E. coti or /f-lactoglobulin. but presumably not to poliovirus which is only exceptionally present in the milieux. They could also be a result of the previously suggested dysregulation of antibody responses in IgAd. Professor Lars A. Hanson. Deparlmem of Clinical Immunology. University of Goteborg, Guldheihgatan 10 A. S-4I3 46 Goieborg. Sweden
IgA deficiency (IgAd) is a common but heterogeneous condition. Although many individuals with IgAd are healthy, others are affected by various disorders such as malabsorption, atopy, autoimmune diseases, tumours and recurrent infections, mainly involving Ihc upper respiratory tract [1]. Some of these symptoms, e.g. the increased frequency of infections, may be related to the antibody defieieney since the lack of serum IgA is generally accompanied by a defieieney of secretory IgA which is a major component of the mucosa! defence system. However, it is still not clear why many individuals with IgAd do not present with any symptoms despite their antibody deficiency. Some of them seem to compensate for their antibody defect with an increased concen-
tration of IgM antibodies in secretions and they have been found to have an increased number of IgM-producing immunocytes in their nasal mucosa [2]. Previous studies have shown that quite frequently IgAd is also associated with IgG subelass deficiency [3]. In some reports these cases of combined defieieney seem to be more seriously affected by respiratory infections and may have impaired lung function [4], although this has not been seen in other materials [5]. Little is known about the qualitative characteristics of antibody responses in IgAd. The biological effieacy of antibodies is related to several characteristics, including avidity which depends on antibody affinity and antigen valeney. Avidity can be evaluated in vitro by different methods.
279
280
F- Cardinale el al.
for example by the use of monoclonal antibodyhapten equilibrium dialysis teehniques or the ammonium sulphate radioimmunoprecipitation or enzyme-linked immunosorbent dissociation assays. A method has been described reeently by Pullen el al. for the measurement of isotypcspeeific relative antibody avidity in antibody mixtures employing the thiocyanate chaotropic ion to elute antibodies from solid phase immobilized antigens [6]. The results eompared well with those obtained with equilibrium dialysis. In the present study we measured the antibody levels and the relative avidity of serum IgM and IgG antibodies to bacterial, viral and food antigens in adult IgAd patients with frequent respiratory tract infeetions and healthy IgAd individuals in comparison with healthy IgA replete people.
MATERIALS AND METHODS Subjects. Blood samples were taken from 33 IgAd (serum IgA < 0.05 g/l) patients (8 males and 25 females) affected by respiratory tract infeetions. In addition 18 heailhy IgAd individuals were included (17 males and I female). Fifteen healthy individuals (5 males and 10 females) with normal immunoglobulin levels were used as controls. Sera were stored al —20 C in plastic tubes until used. IgA deficiency was defined as serum IgA less than 0.05 g/l and IgM. IgG and IgGl 4 above the lower limit of the normal range; 0.5; 7.0; 4.22; 1. 17; 0.41 and 0.01 g/l respectively [1,7]. Patients with at least three respiratory tract infections per year requiring aniibiotic treatment and/or with respiratory traet infection symptoms for more than 100 days per year were defined as having frequent infections. All the palients had been referred to our hospital because of their infections. The healthy IgA deficient individuals were found when screening blood donors. None of the individuals included in the study had received any blood transfusion or immunoglobulin infusion during the 3 months prior to blood sampling. Antigen preparation and dilution. E.stherichia coli O antigens from the 10 most common O types were prepared as described previously and pooled [8]. Optimal dilution in phosphate-buffered saline pH 7.2 (PBS), was used for coating. Poliovirus type I antigen was kindly provided by Dr Osterhaus at the Rijksinstitut voor Volksgesundheid (Holland) and diluted 1/350 in PBS. /?-lactoglobulin (Sigma Chemical Co., St Louis. MO. USA) was used at a concentration of 5/ig/ml PBS. Antibody measurements. IgG and IgM antibody levels were determined by a standard enzyme-linked immunosorbent assay (ELISA) [S]. Results were expresed as "/. of a reference serum. Avidity of IgG and IgM antibodies. Theavidity of IgG and IgM antibodies lo the E. eoli O pool antigen, the poliovirus type 1 antigen and yJ-lactoglobulin was determined by the thiocyanate elution enzyme immunoassay [6]. This method expresses relative avidity as
the thiocyanate molarity required to elute 50% of bound antibody under conditions of antigen excess. Briefly, 100 fi] of the antigen diluted to optimal concentration were added to each well of 96-well microtitre plates (Nunc-Immunoplate I, Copenhagen, Denmark) and incubated overnight at room temperature (20 C). After washing the plates the samples were added at a dilution giving an absorption of approximately 0.6 after 100 min reaction time with the substrate and incubation for 4 h in the antigen-coated plates. After four washes with PBS containing 0.05% Tween20 (PBS-T), 100 ;il of KSCN in PBS-T of appropriate molarities (O.I, 0.25, 0.5. 1,0, 2.0. 3.0 and 4.0 M) were added in duplicate wells to elute antigen-bound antibodies. Two replicate wells for each sample were ineubated with PBS-T alone on the same plate. Plates were then incubated for 30 min at room temperature. After six washes wilh PBS-T. 100 ^l of affinitypurified rabbit anli-human IgG, or IgM (Dakopatts Immunoglobulin AG. Copenhagen, Denmark) conjugated to alkaline phosphatase (Boehringer-Mannheim. Mannheim, Germany) and diluted to optimal concentrations in PBS-T for IgM and IgG respectively, were added to each well. Plates were then ineubated overnight at room temperature and washed four times with PBS-T. After adding 100 n] per well of 1 mg/ml crystalline p-nitrophenyl-phosphate disodium in 1 M diethanolamineconlaining 0.002 M MgCI;(pH 9.8). the plates were incubaled in the dark at room temperature fora maximum of 100min. Absorbances were read at a wavelength of 405 nm in a Titertek Multiskan MCC/ 340 (Flow Laboratories. Solna, Sweden). The mean absorbance values for the duplicate wells were plotted after subtraction of the sample-free background values. The relative avidity was expressed as the molarity of KSCN giving 50% of the absorbance for the duplicate wells obtained in the absence of thioeyanale. Statistics. Results are presented as means+ SEM. The statistical analysis was made with the Kruska!Wallis K-sample lest and the Mann-Whitney O'-test. P values less than 0.01 were considered as signifieanl. Only two-tailed tests were used.
RESULTS The level of IgM antibodies against the bacterial, viral and food antigens did not differ between the IgAd individuals with a history of recurrent infections and IgA replete individuals used as controls (Fig. ta-c). However, the IgM antibody levels against both the microbes and food antigen were significantly higher in the IgAd group with infections compared with the healthy IgAd individuals (Fig. la-c). The level of IgG antibodies against the viral and food antigen did not differ significantly between the three groups, but the IgG antibody levels against E. coti O antigens were signifieantly higher in the symptomatic IgAd individuals eompared with the healthy IgAd group and controls.
Serum IgM and IgG Antibodies in IgA Deficiency
281
P
Aberrations in Titre and Avidity of Serum IgM and IgG Antibodies to Microbial and Food Antigens in IgA Deficiency F. CARDINALE*§, V. FRIMANf. B. CARLSSON*, J. BJORKANDERJ L. ARMENIO§ & L. A. HANSON* Departments of *ClinicaI Immutiology, tinfectious Diseases and Jlnternal Medicine I, The Asthma and Allergy Research Center. University of Goteborg, Goteborg, Sweden, and §Department of Clinical and Social Paediatrics, University of Bari, Bari, Italy
Cardinale F, Friman V. Carlsson B, Bjorkander J, Armenio L. Hanson L A . Aberrations in Titre and Avidity of Serum IgM and IgG Aniibodies to Microbial and Food Antigens in IgA Deficiency. Scand J (mmunol l992;36:279-83 The antibody levels and relative avidity of serum IgM and IgG aniibodies against E. coli O antigens, poliovirus type 1 and /J-lactoglobulin were determined with enzyme-linked immunosorbent techniques in IgA deficient (IgAd) patients with frequent respiratory tract infections and healthy IgAd individuals. Heallhy individuals with normal tmmunoglobulin levels served as controls. The IgM antibody levels against the bacterial, viral and food antigens and the IgG antibody levels against the bacterial antigens were significantly higher in the IgAd group with recurrent infections than in the group of healthy IgAd individuals. The symptomatic IgAd group had significantly higher levels of the IgG antibodies against the bacterial antigen, also when eompared with controls. In contrast the healthy IgAd individuals had the highesi avidities of IgM antibodies to the viral and food antigens. The high avidities of aniibodies could be a compensatory host defence mechanism in IgAd. These aberrations may appear as a consequence of increased mucosal exposure in IgAd to antigens such as E. coti or /f-lactoglobulin. but presumably not to poliovirus which is only exceptionally present in the milieux. They could also be a result of the previously suggested dysregulation of antibody responses in IgAd. Professor Lars A. Hanson. Deparlmem of Clinical Immunology. University of Goteborg, Guldheihgatan 10 A. S-4I3 46 Goieborg. Sweden
IgA deficiency (IgAd) is a common but heterogeneous condition. Although many individuals with IgAd are healthy, others are affected by various disorders such as malabsorption, atopy, autoimmune diseases, tumours and recurrent infections, mainly involving Ihc upper respiratory tract [1]. Some of these symptoms, e.g. the increased frequency of infections, may be related to the antibody defieieney since the lack of serum IgA is generally accompanied by a defieieney of secretory IgA which is a major component of the mucosa! defence system. However, it is still not clear why many individuals with IgAd do not present with any symptoms despite their antibody deficiency. Some of them seem to compensate for their antibody defect with an increased concen-
tration of IgM antibodies in secretions and they have been found to have an increased number of IgM-producing immunocytes in their nasal mucosa [2]. Previous studies have shown that quite frequently IgAd is also associated with IgG subelass deficiency [3]. In some reports these cases of combined defieieney seem to be more seriously affected by respiratory infections and may have impaired lung function [4], although this has not been seen in other materials [5]. Little is known about the qualitative characteristics of antibody responses in IgAd. The biological effieacy of antibodies is related to several characteristics, including avidity which depends on antibody affinity and antigen valeney. Avidity can be evaluated in vitro by different methods.
279
280
F- Cardinale el al.
for example by the use of monoclonal antibodyhapten equilibrium dialysis teehniques or the ammonium sulphate radioimmunoprecipitation or enzyme-linked immunosorbent dissociation assays. A method has been described reeently by Pullen el al. for the measurement of isotypcspeeific relative antibody avidity in antibody mixtures employing the thiocyanate chaotropic ion to elute antibodies from solid phase immobilized antigens [6]. The results eompared well with those obtained with equilibrium dialysis. In the present study we measured the antibody levels and the relative avidity of serum IgM and IgG antibodies to bacterial, viral and food antigens in adult IgAd patients with frequent respiratory tract infeetions and healthy IgAd individuals in comparison with healthy IgA replete people.
MATERIALS AND METHODS Subjects. Blood samples were taken from 33 IgAd (serum IgA < 0.05 g/l) patients (8 males and 25 females) affected by respiratory tract infeetions. In addition 18 heailhy IgAd individuals were included (17 males and I female). Fifteen healthy individuals (5 males and 10 females) with normal immunoglobulin levels were used as controls. Sera were stored al —20 C in plastic tubes until used. IgA deficiency was defined as serum IgA less than 0.05 g/l and IgM. IgG and IgGl 4 above the lower limit of the normal range; 0.5; 7.0; 4.22; 1. 17; 0.41 and 0.01 g/l respectively [1,7]. Patients with at least three respiratory tract infections per year requiring aniibiotic treatment and/or with respiratory traet infection symptoms for more than 100 days per year were defined as having frequent infections. All the palients had been referred to our hospital because of their infections. The healthy IgA deficient individuals were found when screening blood donors. None of the individuals included in the study had received any blood transfusion or immunoglobulin infusion during the 3 months prior to blood sampling. Antigen preparation and dilution. E.stherichia coli O antigens from the 10 most common O types were prepared as described previously and pooled [8]. Optimal dilution in phosphate-buffered saline pH 7.2 (PBS), was used for coating. Poliovirus type I antigen was kindly provided by Dr Osterhaus at the Rijksinstitut voor Volksgesundheid (Holland) and diluted 1/350 in PBS. /?-lactoglobulin (Sigma Chemical Co., St Louis. MO. USA) was used at a concentration of 5/ig/ml PBS. Antibody measurements. IgG and IgM antibody levels were determined by a standard enzyme-linked immunosorbent assay (ELISA) [S]. Results were expresed as "/. of a reference serum. Avidity of IgG and IgM antibodies. Theavidity of IgG and IgM antibodies lo the E. eoli O pool antigen, the poliovirus type 1 antigen and yJ-lactoglobulin was determined by the thiocyanate elution enzyme immunoassay [6]. This method expresses relative avidity as
the thiocyanate molarity required to elute 50% of bound antibody under conditions of antigen excess. Briefly, 100 fi] of the antigen diluted to optimal concentration were added to each well of 96-well microtitre plates (Nunc-Immunoplate I, Copenhagen, Denmark) and incubated overnight at room temperature (20 C). After washing the plates the samples were added at a dilution giving an absorption of approximately 0.6 after 100 min reaction time with the substrate and incubation for 4 h in the antigen-coated plates. After four washes with PBS containing 0.05% Tween20 (PBS-T), 100 ;il of KSCN in PBS-T of appropriate molarities (O.I, 0.25, 0.5. 1,0, 2.0. 3.0 and 4.0 M) were added in duplicate wells to elute antigen-bound antibodies. Two replicate wells for each sample were ineubated with PBS-T alone on the same plate. Plates were then incubated for 30 min at room temperature. After six washes wilh PBS-T. 100 ^l of affinitypurified rabbit anli-human IgG, or IgM (Dakopatts Immunoglobulin AG. Copenhagen, Denmark) conjugated to alkaline phosphatase (Boehringer-Mannheim. Mannheim, Germany) and diluted to optimal concentrations in PBS-T for IgM and IgG respectively, were added to each well. Plates were then ineubated overnight at room temperature and washed four times with PBS-T. After adding 100 n] per well of 1 mg/ml crystalline p-nitrophenyl-phosphate disodium in 1 M diethanolamineconlaining 0.002 M MgCI;(pH 9.8). the plates were incubaled in the dark at room temperature fora maximum of 100min. Absorbances were read at a wavelength of 405 nm in a Titertek Multiskan MCC/ 340 (Flow Laboratories. Solna, Sweden). The mean absorbance values for the duplicate wells were plotted after subtraction of the sample-free background values. The relative avidity was expressed as the molarity of KSCN giving 50% of the absorbance for the duplicate wells obtained in the absence of thioeyanale. Statistics. Results are presented as means+ SEM. The statistical analysis was made with the Kruska!Wallis K-sample lest and the Mann-Whitney O'-test. P values less than 0.01 were considered as signifieanl. Only two-tailed tests were used.
RESULTS The level of IgM antibodies against the bacterial, viral and food antigens did not differ between the IgAd individuals with a history of recurrent infections and IgA replete individuals used as controls (Fig. ta-c). However, the IgM antibody levels against both the microbes and food antigen were significantly higher in the IgAd group with infections compared with the healthy IgAd individuals (Fig. la-c). The level of IgG antibodies against the viral and food antigen did not differ significantly between the three groups, but the IgG antibody levels against E. coti O antigens were signifieantly higher in the symptomatic IgAd individuals eompared with the healthy IgAd group and controls.
Serum IgM and IgG Antibodies in IgA Deficiency
281
P
