A mechanism for the induction of immunological tolerance by antigen feeding: antigen-antibody complexes.

We have previously reported on the induction, in mice, of a systemic (splenic) immune response with IgA as the dominant antibody, as a result of a short (4 day) intragastric immunization course with foreign erythrocytes. This response was followed by a prolonged period of hyporesponsiveness to similarly administered antigen. Here it is shown that this hyporesponsiveness is also manifested towards antigen given intraperitoneally, and that one is therefore dealing with tolerance, not with failure to absorb antigen from the gut. In contrast, mice primed parenterally and then challenged intragastrically behaved as if never having any previous contact with the antigen, i.e., with a primary-type splenic response of predominant IgA character. This agrees with our former conclusion that splenic responses to enterically absorbed antigen reflect colonization of the spleen by cells sensitized locally in the gut wall, a site not readily primed by the parenteral route. Serum from intragastrically immunized mice contained a very active tolerogen. In vivo, it was capable of conferring tolerance to nonimmune recipient mice. In vitro, it paralyzed the activity of antibody-producing cells. Inhibitory sera has weak antibody activity, restricted to the IgA class, and contained immune complexes reacting with rheumatoid factor but not with C1q. Elimination of these complexes by means by insolubilized rheumatoid factor abolished the tolerogenic effect. In conclusion, the enterically induced tolerogen seems to consist of immune complexes with IgA as the antibody.

High levels of secretory IgA and free secretory component in the serum of rats with bile duct obstruction.

In the rat , ligation of the bile duct induces a rapid and progressive elevation of the IgA levels in serum. The increase is about 4-fold at 1 h, 15-fold at 1 day, and 30-fold at 1 wk after ligation. The additional IgA is of the secretory type. Free secretory component also appears in serum after bile duct obstruction; it does not continue to increase and occasionally disappears from serum after prolonged ligation. The increase in serum IgA levels is selective. These changes are totally reversible if the bile duct is reopened at 1 day after ligature. These findings confirm the role of the rat liver in the transfer of circulating IgA into the bile.

Subclasses of human immunoglobulin a based on differences in the alpha polypeptide chains.

Antiserum from goats immunized with heavy polypeptide chains from a gammaA-type myeloma globulin was absorbed with serum from patients with selective absence of immunoglobulin A (gammaA). The resulting reagents could be used for the classification of 58 gammaA-myeloma proteins into two distinct antigenic types, respectively called subclasses He and Le. These differences were shown to be related to the heavy (alpha) polypeptide chains and independent of the integrity of interchain disulfide bridges. The gammaA-immunoglobulin from normal serum appears to consist, for the most part, of molecules with Le subclass specificity.

Selective transport of polymeric immunoglobulin A in bile. Quantitative relationships of monomeric and polymeric immunoglobulin A, immunoglobulin M, and other proteins in serum, bile, and saliva.

In 17 adults, serum, hepatic bile, and saliva samples were analyzed for their sedimentation profile of IgA and secretory component (SC), and for their concentrations of albumin, orosomucoid, transferrin, IgG, IgA, alpha 2-macroglobulin (alpha 2M), IgM, and SC. Polymeric IgA(p-IgA) averaged 13% (50-700 micrograms/ml) of total IgA in serum, 70% (43-88%) in bile, and 93% (74-98%) in saliva. Most of the p-IgA in bile sedimented with SC, which also occurred free (8-44%), and with IgM. In bile, albumin (155-1,485 micrograms/ml) was the predominant protein, followed by IgG (32-480 micrograms/ml), and total IgA (37-209 micrograms/ml). In saliva, p-IgA (72-902 micrograms/ml) predominated, followed by albumin (16-385 micrograms/ml) and IgG (9-178 micrograms/ml). Secretion-to-serum albumin-relative concentration ratios (S/S-ARCR = 1 for albumin) in bile averaged 22 for p-IgA, 1.91 for IgM, 1.28 for monomeric IgA (m-IgA), 0.70 for IgG, and 0.57 for alpha 2M, indicating for p-IgA, IgM, and to a lesser extent for m-IgA, a selective excretion into bile. In saliva, a 16-fold greater selective excretion of p-IgA (mean S/S-ARCR = 354) was found. Labeled m- and p-IgA were injected intravenously into five patients. Specific activities indicated that for p-IgA 50% was serum derived in bile, as compared with 2% in saliva, and to 85% for m-IgA in bile. In the patient with the highest excretion of 125I-p-IgA in bile, only 2.8% of the injected dose was recovered in bile within 24 h after injection. Compared with rats and rabbits, the serum-to-bile transport of p-IgA in humans is much smaller.

Changes in size, subclass, and metabolic properties of serum immunoglobulin A in liver diseases and in other diseases with high serum immunoglobulin A.

We have studied the relative contributions of monomeric (m-) and polymeric IgA (p-IgA) and of IgA1 and IgA2 to total serum IgA in healthy adults and patients with liver disease (LD) or with other diseases and high serum IgA. Serum concentration of total secretory component (SC) was also determined. In addition, fractional catabolic rates (FCR) and synthetic rates for both m- and p-IgA were measured in nine controls and nine cirrhotics. Our results support four main conclusions: (a) In healthy adults, intravascular p-IgA contributes to only 4-22% (mean 12%) of serum IgA, because its FCR and synthetic rate are approximately two times higher and four times smaller, respectively, than those of intravascular m-IgA. (b) in LD, biliary obstruction does not result in a significant increase in serum p-IgA unlike in rats and rabbits, indicating that in humans the SC-dependent biliary transport of p-IgA plays a much less significant role in selective removal of p-IgA from plasma than in rats and rabbits. (c) In contrast to biliary obstruction, parenchymal LD results in a significant and preferential increase in serum p-IgA, which in cirrhotics correlates with a selective reduction of the p-IgA-FCR. This supports a role for the human liver in selective removal of p-IgA from plasma, but another mechanism than the SC-dependent biliary transport should be considered. (d) Total SC, p-IgA, and IgA2 in serum are unlinked parameters, not necessarily reflecting mucosal events. A marked increase in serum SC occurs almost selectively in LD. Although a shift to IgA2 is suggested in Crohn's disease and alcoholic cirrhosis, a shift to IgA1 frequently associated to a shift to p-IgA occurs in chronic active LD, primary Sicca, and connective tissue diseases.

Secretion of immunoglobulins and plasma proteins from the jejunal mucosa. Transport rate and origin of polymeric immunoglobulin A.

Parameters of secretion of IgA and several other plasma proteins from the jejunal mucosa were investigated in 11 individuals who had a normal distribution of Ig-containing cells in the lamina propria and in one patient who was totally deficient in jejunal IgA and IgM plasmacytes. Jejunal samples were collected during segmental gut perfusion. The following results were obtained: (a) The secretion of polymeric IgA (p-IgA, mean equals 217 micrograms/40 cm per min) exceeded those of albumin (132 micrograms), IgG (35 micrograms), and monomeric IgA (m-IgA, 15 micrograms, or 6.4% of total IgA). About 35% of IgA was IgA2 in the jejunal secretion, compared with approximately 23% in serum. This closely corresponds to the 35 and 24% of IgA2 plasmocytes in jejunal mucosa and peripheral lymph nodes, respectively. (b) For each protein, a relative coefficient of excretion (RCE) was calculated (jejunum to serum concentration ratio expressed relative to that of albumin). RCEs of 1.41 for orosomucoid, 1.0 for albumin, 0.83 for IgG, and 0.74 for IgE and, in the deficient patient, of 0.64 for m-IgA and 0.016 for IgM were obtained. This was inversely related to the molecular weight of these proteins and indicated their predominantly passive transport into the jejunum. However, in normal individuals, the RCE of transferrin (approximately 1.11 greater than 1, P greater than 0.05), alpha 2-macro globulin (approximately 0.77), m-IgA (approximately 1.98), and p-IgA (approximately 218) exceeded the value expected from simple seepage from plasma, thus pointing to an additional role of either local gut synthesis and/or active transepithelial transport. (c) Approximately 98% of p-IgA, approximately 99% of IgM, and approximately 68% of m-IgA in jejunal secretions were derived from local production in the gut wall, as determined by 125I-p-IgA specific activities and/or by comparison between the RCE values of the deficient patient to the values of controls. Therefore, the jejunal production of p-IgA (approximately 312 mg/d per 40 cm vs. approximately 54 mg/d from bile) contributes the majority of upper intestinal IgA in humans. The active transport of plasma p-IgA across the intestinal mucosa (approximately 0.08 mg/40 cm per kg per d) contributes less than 2% of the total amount of p-IgA (4.5 mg/kg per d) that is cleared daily from plasma.

Esterase activity associated with secretory IgA and free secretory component preparations from human milk.

Secretory immunoglobulin A (IgA) and free secretory component were purified from human milk using methods involving (NH4)2SO4 precipitation, ion-exchange and gel-filtration chromatography. It was found that the preparations, although apparently pure by conventional criteria based on immunoelectrophoresis, zone electrophoresis, or ultracentrifugation, were consistently contaminated by two different esterases. These enzymes could apparently not be removed by further ion-exchange chromatography and/or gel filtration. They could, however, be eliminated by passage on an immunoabsorbent made from anti-free secretory component antiserum.

Influence of molecular size of IgA on its immunoassay by various techniques--I. Direct and reversed single radial immunodiffusion.

Immunochemically pure samples of monoclonal and polyclonal IgA were prepared from human sera and milk. Samples of various homogeneous molecular sizes were further obtained by preparative ultracentrifugations. The different behaviour of each preparation (monomer, dimer, trimer, tetramer and secretory IgA) were studied in direct (DRID) and reversed (RRID) single radial immunodiffusion using three different anti-alpha-chain antisera and IgA samples of various monoclonal and polyclonal origins. In DRID, all polymer concentrations were underestimated when using monomers as standards, on an equal weight (OD) basis. Correction factors (CFs) were calculated from monomer to polymer DRID slope ratios. The means +/- SDs of these CFs were 1.55 +/- 0.18 for serum dimers, 1.85 +/- 0.10 for trimers, 2.63 +/- 0.26 for tetramers and 2.24 +/- 0.15 for secretory IgA (84% 11S, 16% 15S). These results were confirmed by RRID.

Jacalin: isolation, characterization, and influence of various factors on its interaction with human IgA1, as assessed by precipitation and latex agglutination.

An IgA1-specific lectin, Jacalin, was isolated from dried seeds of the jackfruit, Artocarpus integrifolia, by affinity binding to IgA1-Sepharose and elution with D-galactose. Jacalin is a glycoprotein with two non-covalently bound subunits (15 and 18 K). Interactions between Jacalin and human Igs were studied by precipitation in gel and in solution, and by agglutination of IgA1-coated latex by Jacalin. Jacalin precipitated only with IgA1-containing samples, including monomers, polymers, monoclonal, polyclonal and secretory IgA1, but not IgA2 of both A2m(1) and A2m(2) allotypes, nor with IgG1, 2, 3 and 4, IgM, IgD, and IgE; after neuraminidase treatment, only IgA1 and IgD were precipitated. Jacalin had a relatively broad pH range of activity in both precipitation and agglutination of IgA1-latex. Bivalent metal cations (Ca, Mg, Mn, Cu, Zn, Co, Cd), EDTA, Triton X-100, Tween-20, Na deoxycholate and ionic strength did not influence these reactions. Na dodecylsulphate, guanidine and urea inhibited the reactions whereas NP-40 rather enhanced them. Among 39 types of sugar tested, 10 displayed inhibitory activity, decreasing in the following order: p-nitrophenyl-alpha-D-galactopyranoside, 1-O-methyl-alpha-D-galactopyranoside, D-melibiose, p-nitrophenyl-beta-D-galactopyranoside, GalNAc, stachyose, 1-O-methyl-alpha-D-mannopyranoside, D-galactose, D-galactosamine and 1-O-methyl-alpha-D-glucopyranoside. IgA1, treated with neuraminidase or not, but not the other human Igs, was also an excellent inhibitor of agglutination, being more powerful than the best sugars studied. Only neuraminidase-treated IgD was also inhibitory, but less so than IgA1. Jacalin preferentially bound to alpha-linked non-reducing D-galactose. The configuration of OH-groups at C-2, C-4 and C-6 of D-galactose was important for the reaction. Jacalin recognizes terminal Gal beta 1-3GalNac-, as in the IgA1-hinge, and/or GalNAc-, but not Gal beta 1-4GlcNAc-, nor Gal beta 1-6GlcNAc-, nor their sialylayted extensions. Latex agglutination and its inhibition assay are particularly well suited for the study of these lectin-glycoprotein interactions.

IgA-induced chemokinesis of human polymorphonuclear neutrophils: requirement of their Fc-alpha receptor.

The influence of purified human immunoglobulins on the migration of human neutrophils (PMN) was measured in a 48-well micro chemotaxis chamber, with results expressed as percentages of maximal formyl-methionyl-leucyl-phenylalanine (FMLP)-stimulated chemotaxis. Both monomeric and polymeric IgA, of both subclasses, in monoclonal and polyclonal form, as well as secretory IgA and Fc-alpha, but not Fab-alpha fragments, enhanced PMN migration when present either in the lower or in both compartments of the chamber (chemokinesis) at concns as low as 0.1 mg/ml. IgM and IgE had no such effect. In contrast, IgG was chemotactic at low concn (0.1 mg/ml). Both monomeric and polymeric IgA decreased the maximally induced FMLP-chemotaxis, but IgA increased chemotaxis induced by suboptimal levels of FMLP. Binding of 3[H]-FMLP to PMN was not affected. Cytofluorographic analysis revealed that, under the conditions of the assay, IgA did bind to 93% of PMN. Thus, the various forms of IgA have a dual effect on human PMN mobility: (1) increase PMN random migration (chemokinesis); and (2) decrease the maximal FMLP-induced chemotaxis. Our data support the requirement of binding of IgA to the Fc-alpha receptor of PMN for expression of these activities. This effect of IgA on PMN mobility may be relevant in IgA deficiency states.

Interaction between streptococcal protein Arp and different molecular forms of human immunoglobulin A.

Protein Arp, the IgA-binding protein of the group A Streptococcus, has affinity for the Fc-part of IgA. The binding between protein Arp and several different molecular forms of human IgA was characterized. It was found that protein Arp bound with higher affinity to uncomplexed forms of IgA than to complexed forms (secretory IgA, alpha 1-antitrypsin-IgA and alpha 1-microglobulin-IgA). Thus, the affinity constant was 2.0-5.9 x 10(8) M-1 for the binding to monomeric, dimeric, trimeric, and quadrimeric IgA, and 4.5-5.0 x 10(7) M-1 for binding to the complexed forms. Among the uncomplexed IgA-molecules, the affinity constant was in the same range for J chain-containing forms (dimeric, trimeric and quadrimeric IgA) as for forms without J chain (monomeric and a particular quadrimeric IgA devoid of J chain). Western blotting demonstrated that protein Arp bound exclusively to the alpha-chain of all IgA-forms. Several lines of evidence pointed to a localization of the binding site to the C alpha 3-domain. First, protein Arp did not bind to three N-terminal alpha-chain fragments which lacked a region corresponding to the C alpha 3-domain, including that form a four-chain myeloma IgA, naturally occurring in plasma. Second, the binding to dimeric and tri/quadrimeric IgA was partially blocked by an added secretory component, which has been suggested to bind to the C alpha 2- and C alpha 3-domains of the alpha-chain. Finally, alpha 1-antitrypsin and alpha 1-microglobulin, in the weakly binding IgA-complexes, have been shown to be linked to the C alpha 3-domain via the penultimate amino acid residue of the alpha-chain peptide, supporting the hypothesis of a localization of the binding site of protein Arp to the C alpha 3-domain.

Complement-mediated solubilization of rat IgA immune precipitates.

The complement-mediated solubilization (CMS) of immunoprecipitates (IP) consisting of DNP-rat serum albumin (RSA) and rat monoclonal anti-DNP antibodies of the IgA [both polymeric (p-) and monomeric (m-)] or IgG2b (sub)class was studied. In contrast to IgG2b IP, solubilization of IgA IP was only observed in an autologous system, with rat serum as the source of complement. IP prepared using m-IgA were solubilized faster than those prepared using p-IgA. Analysis of both affinity and avidity of the antibodies, indicated that this difference may be due to the lower avidity of the m-IgA antibodies as compared to p-IgA. Analysis of the solubilized IP revealed deposition of C3 and C4 on IgG2b, and only C3 on IgA IP. These results point toward a role of the alternative pathway in the solubilization of IgA IP. Size analysis of the solubilized IgA IP employing sucrose density gradient ultracentrifugation, indicated that these were heterogeneous, with a size generally larger than 19 S.

Rat polymeric IgA binds C1q, but does not activate C1.

Immune complexes, prepared with monoclonal rat IgA antibodies directed against DNP, activate the alternative pathway of the complement system in rat serum. In this study, the interaction of these monoclonal IgA antibodies with the classical pathway of complement was investigated. Monoclonal polymeric IgA (p-IgA) was shown to inhibit the IgG2b-mediated classical pathway-dependent lysis of TNP-coated sheep red blood cells. In addition, the binding of C3 to solid phase IgG2b immune complexes was inhibited by p-IgA. Monoclonal monomeric IgA (m-IgA) was much less efficient in this respect. To further analyse the effect of p-IgA on the activation of the classical pathway by IgG2b immune complexes, the interaction of p-IgA with C1 was studied. It was found that p-IgA antibodies bind C1q. No species-specificity was observed, since both rat and human C1q were bound. Whereas binding of C1q in C1 to IgG2b resulted in activation of C1, binding to p-IgA did not. The binding of C1q to both p-IgA and IgG2b could be inhibited by monoclonal antibodies directed against the globular heads of C1q, but not by monoclonal antibodies directed against the collagen tail. The formation of insoluble p-IgA immune complexes was inhibited in the presence of rat serum or C1. These studies indicate that C1q binds to p-IgA by its globular heads, and thereby may modulate classical pathway-mediated reactions such as the inhibition of immune precipitate formation.

Monoclonal antibodies against isotypic and isoallotypic determinants of human IgA1 and IgA2: fine specificities and binding properties.

We have analysed and compared the fine specificity and behavior in various immunoassays of 10 mouse monoclonal antibodies, from three independent laboratories, directed against IgA1, IgA2 or non-IgA2m(2). The following observations were made. (1) Although all of the monoclonal antibodies were specific for a particular IgA subclass or isoallotype in a radioimmunoassay, three of them were not specific when tested in indirect immunofluorescence on plasma cells derived from pokeweed-activated peripheral blood lymphocytes. In this highly sensitive system, contrary to direct immunofluorescence previously performed using formalin-fixed lymphoid tissue, the anti-IgA1 69.114 reacted with some of the IgA2 plasma cells, the anti-IgA2 DLDB7 reacted with some of the IgA1 plasma cells and the anti-IgA2 16.512 dimly reacted with all IgM plasma cells. (2) Among the eight anti-IgA subclass antibodies, seven were directed against the CH2 domain of IgA whereas the anti-IgA1 1-155-1 recognised an epitope destroyed by Streptococcus sanguis IgA1 protease and localised in the hinge region of IgA1. The two anti-isoallotype antibodies were directed against epitope(s) probably localised in the 65 C-terminal amino acid residues of the alpha-CH3 domain. All of the 10 antibodies were able to react with endogeneously produced surface IgA on B-cells. (3) Using monoclonal anti-IgA subclass antibodies in radioimmunoassay may be hazardous in the absence of knowledge of their affinity constants and of careful control experiments: some of the antibodies were not sensitive in radioimmunoassays designed to measure the serum titer of specific IgA1 and IgA2 antibodies. Moreover, major differences were observed between the different monoclonal reagents with respect to the influence of the size of IgA on a solid-phase sandwich radioimmunoassay. While three of the anti-IgA1 underestimated dimeric IgA relative to monomeric IgA, the fourth anti-IgA1 and all the anti-IgA2 overestimated dimeric IgA relative to monomeric IgA, by a factor sometimes close to 7.

Influence of molecular size of IgA on its immunoassay by various techniques. III. Immunonephelometry.

The influence of size heterogeneity of IgA on its immunonephelometric (IN) assay was studied using highly purified preparations of monoclonal (MC) or polyclonal (PC) monomeric (M), dimeric (D), trimeric (T) and tetrameric (Q) IgA, as well as of secretory IgA (sIgA). Concentrations measured by optical density (OD) were compared to IN--concentrations obtained by using M as standards, on an equal weight basis. No significant difference was found between OD and IN concentrations of D, T, Q and sIgA were only slightly underestimated, with correction factors (CF) of 1.16, 1.20 and 1.24, respectively. The value of 1.24 for our sIgA sample (84% of 11S, 16% of 15S) could be due to the fact that secretory component represents about 20% of the mass of the sIgA molecule. IN seems a method of choice for measuring IgA concentrations in the range of microgram/ml, almost independently of the size of IgA.

A solid phase, direct competition, radioimmunoassay for quantitation of secretory IgA in human serum.

A solid phase radioimmunoassay is described for measurement of secretory IgA (sIgA) in human serum. The assay is based on direct competition between sIgA in serum and purified labelled milk sIgA for anti-secretory-component antibodies coated on disposable plastic cups. The assay is relatively rapid, reproducible and of adequate sensitivity. A wide range of sIgA concentrations (5--120 microgram/ml) can be tested with the same dilution of human serum. The arithmetic mean +/- S.D. of the serum sIgA levels of 120 random blood donors (both sexes) was 10.9 +/- 4.6 microgram/ml. Compared to previously published methods and results of quantitation of sIgA in human sera, the present assay is an improvement which should yield interesting data in human hepatic physiopathology.

Non-proteic immunogenic impurity in proteins purified by preparative electrophoresis on Pevikon.

Purified IgA proteins, obtained by zonal electrophoresis on blocks of Pevikon, contained an uncharged, non-proteic, water-soluble macromolecular impurity. The contaminant, which precipitates in half-saturated ammonium sulfate, could be an unknown derivative of polyvinyl alcohol. It was discovered because it induced the synthesis of specific precipitating antibodies in a rabbit used for antibody production against the electrophoresed protein, generating spurious precipitin reactions during its immunochemical analysis. Thorough washing of the Pevikon will substantially reduce the amount of impurity and the danger of obtaining precipitating antibodies against it after injection into animals.

Rat monoclonal antibodies. VI. Production of IgA secreting hybridomas with specificity for the 2,4-dinitrophenyl (DNP) hapten.

A simple method to obtain rat hybridomas producing specific IgA antibodies is reported. By fusing the IR983F rat myeloma cell line with mesenteric lymph node cells from LOU/C rats immunized via the Peyer's patches with DNP-Salmonella typhimurium, twenty hybrids secreting monoclonal IgA antibodies specific for DNP were produced and maintained as highly secreting transplantable ascitic tumors. The monoclonal IgA antibodies were easily purified by affinity chromatography on a DNP-immunosorbent and were found to comprise both monomers and polymers.

Influence of molecular size of IgA on its immunoassay by various techniques. II. Solid-phase radioimmunoassays.

Homogeneous, polymeric (P), monoclonal (MC) and polyclonal (PC) samples of purified IgA, as well as of secretory IgA (sIgA), were compared to their corresponding monomeric (M) forms with regard to their respective behaviour in both a solid-phase double antibody (AB) immunoradiometric assay (IRMA) and a solid-phase competition radioimmunoassay (CRIA). On an identical weight basis, both assays underestimated the P forms. Correction factors (CF), i.e., optical density (OD) versus radiometric concentration ratios, were calculated for both methods for all P forms, using M as standards. IRMA CF were respectively 1.50 for dimer, 1.98 for secretory IgA and 2.40 for tetramers, very similar to those obtained in radial immunodiffusion (RID). With optimally coated beads, these CF were stable along a wide range of concentrations. In contrast, CRIA-CF were 3-4 times larger and displayed much variation along their standard range of concentrations, preventing the use of a constant CF along the standard curve. Our present IRMA, with its CF, should be of value for a more accurate quantitation of small amounts of IgA of various known sizes.

Particle counting immunoassay. IV. The use of F(ab')2 fragments and N epsilon-chloroacetyl lysine N-carboxyanhydride for their coupling to polystyrene latex particles.

Improved performance with the latex agglutination immunoassay, PACIA, is possible when the F(ab')2 fragments of antibody are chemically coupled at their hinge region to a protein-coated latex using a new coupling reagent, N epsilon-chloroacetyl lysine N-carboxyanhydride. This reagent, whether by orientating the complete antibody molecule or its F(ab')2 fragment or by preventing antibody denaturation, increased the sensitivity of the PACIA method. The use of F(ab')2 fragments eliminated most serum interference with the test.