Immune complex-like moieties in immunoglobulin for intravenous use (i.v.Ig) bind complement and enhance phagocytosis of human erythrocytes

Abstract

Treatment with i.v.Ig can, on rare occasions, lead to detrimental effects such as enhanced erythrocyte sequestration and an increase in serum immune complexes with inflammatory sequellae such as exacerbation of glomerular nephritis. In this study, i.v.Ig (Sandoglobin) was examined for complement binding moieties which resemble immune complexes and can mediate the binding of IgG and C'3b to human erythrocytes via CR1 and enhance erythrocyte susceptibility to sequestration. Sephacryl S-200 HR separated i.v.Ig into two fractions: monomeric IgG (74%) and larger complexes of the molecular weight of an IgG dimer or greater (> or = 300 kD) (26%). In the presence of complement, the 'dimers' bound to human erythrocytes, rendering them susceptible to phagocytosis in vitro. Removal of erythrocyte-specific isoantibodies from the i.v.Ig had no effect on 'dimer' binding to the erythrocytes. Monomeric IgG contained virtually no complement-activating, erythrocyte-binding activity. Erythrocyte binding of complement-bearing IgG 'dimers' and subsequent phagocytosis resembles the binding of complement-bearing immune complexes to erythrocyte CR1. Exposure to Factor I leads to the release of complement-bearing IgG 'dimers' from erythrocyte CR1 and to the abrogation of erythrophagocytosis. Binding of complement-bearing IgG 'dimers' to the erythrocyte is blocked by To5, a CR1-specific monoclonal antibody.

Fig. 1

Erythrocytes (O+) exposed to IVIg or immune complexes (IC) in the presence of complement bind IgG, C′3c and C′3d and become susceptible to erythrophagocytosis. O+ erythrocytes were incubated with IVIg or tetanus–anti-tetanus IC and pooled AB serum as a source of complement (C′). In parallel, erythrocytes were incubated with IVIg, IC or C′ alone. Aliquots of these erythrocytes were prepared for flow cytometric detection of membrane-bound IgG, C′3c or C′3d. Erythrocytes incubated with second antibody alone served as controls for non-specific binding (a). Aliquots of these same erythrocytes were washed three times in PBS and added to the phagocytosis assay (b). The results are the average ± s.d. of five donors.

Fig. 2

A+ and B+ erythrocytes exposed, in the presence of complement, to IVIg adsorbed on A+ and B+ erythrocytes bind IgG, C′3c and C′3d and become susceptible to phagocytosis. A+ (a) or B+ (b) erythrocytes were exposed, in the presence or absence of pooled AB serum as a source of complement (C′) to IVIg adsorbed on A+ or B+ erythrocytes, respectively, to C′ alone, to unadsorbed (stock) IVIg or to culture medium (in situ). Erythrocytes were prepared for the flow cytometric detection of membrane-bound IgG, C′3c or C′3d as in Fig. 1. Aliquots of similarly treated erythrocytes were washed and added to a phagocytosis assay (c). The results are the average ± s.d. of four A+ donors and three B+ donors for flow cytometry and three A+ and three B+ donors for the phagocytosis assay. ND, Not done.

Fig. 3

Complexes formed by IVIg and complement resemble complement-bearing immune complexes (IC + C′) in their binding to erythrocytes and in their Factor I-mediated release from erythrocytes. O+ and A+ erythrocytes were incubated with stock IVIg or IVIg adsorbed on A+ erythrocytes, respectively, in the presence of pooled AB serum as a source of complement (C′). O+ erythrocytes were incubated with tetanus–anti-tetanus immune complexes (IC) and C′. A+ erythrocytes were incubated with IgM anti-blood group A isoantibodies. Aliquots of these cells were washed and incubated with EDTA serum as a source of Factor I. All erythrocytes were washed and prepared for the flow cytometric detection of membrane-bound IgG (a), C′3c (b) or C′3d (c) as in Fig. 1. The results are the average ± s.d. of four O+ donors and three A+ donors.

Fig. 4

Factor I-mediated release of complexes formed by IVIg and complement abrogates susceptibility of erythrocytes to phagocytosis. A+ erythrocytes were incubated with IVIg (adsorbed on A+ erythrocytes) in the presence of pooled AB serum as a source of complement or with tetanus–anti-tetanus immune complexes (IC) and C′. Aliquots of these erythrocytes were then washed and incubated with serum ETDA. All erythrocytes were rewashed and added to a phagocytosis assay. The results are the average ± s.d. of three donors.

Fig. 5

Blockade of CR1 inhibits the binding to erythrocytes of complexes formed by IVIg and complement. O+ erythrocytes were incubated with the MoAb To5 which blocks the C′3b binding site on CR1 or with the isotype-matched, non-blocking CR1-specific MoAb E11, and then exposed to either IVIg + C′ or IC + C′. A+ erythrocytes were incubated with To5 and IgM anti-A isoagglutinins + C′. Erythrocytes were then washed and prepared for the flow cytometric detection of membrane-bound IgG, C′3c or C′3d as in Fig. 1. The results are the average ± s.d. of three O+ donors and two A+ donors.

Fig. 6

Sephacryl S-200 HR separates IVIg into two fractions of approximate mol. wt 150 kD and ≥ 300 kD. IVIg was separated on Sephacryl S-200 HR. Molecular weight markers ranged from 29 000 to 443 000 D. Fractions were collected and assayed for protein concentration by the BioRad Protein Assay. Protein concentration is expressed in arbitrary units. The peaks marked I and II were pooled for subsequent analysis. The results are of a representative separation of six performed.

Fig. 7

The immune complex-like forms in IVIg capable of activating complement and binding to erythrocyte CR1 are found in the higher molecular weight (≥ 300 kD) Sephacryl S-200 HR fraction of IVIg (Fraction I). Erythrocytes were treated with stock IVIg or with the IVIg Sephacryl fractions I (mol. wt ≥ 300 kD) or II (mol. wt ≈ 150 kD) in the presence of pooled AB serum as a source of complement (C′). Erythrocytes were prepared for the flow cytometric detection of membrane-bound IgG, C′3c or C′3d as in Fig. 1. Specific activity was calculated by dividing the net mean fluorescence of each erythrocyte sample by the amount of protein in the stock IVIg or the appropriate Sephacryl fraction. The results are of a representative experiment of three performed with O+ erythrocytes.