Some theoretical considerations regarding the effects of steric hindrance and intrinsic global coupling on the flexibility of Fc-anchored immunoglobulins

Nanosecond fluorescence depolarization studies reported in the accompanying companion paper showed that the long rotational correlation time, phi L, increased somewhat when rabbit IgG anti-dansyl antibodies were anchored in staphylococcal protein A (SpA) soluble complexes. The increases in phi L upon anchoring IgG probably resulted from "global coupling" effects caused by: increased steric hindrance of the antibody segments in the SpA complexes and intrinsic structural constraints already present in the monomeric IgG. Global coupling results from a restriction in the angular range of a flexible segment and is manifest when flexible motions alone cannot depolarize all of the fluorescence, so that the slower global tumbling of the entire particle is also required. Such effects cannot be resolved directly from experimental anisotropy data, however, because only a single long correlation time, phi L, is well defined over the limited time range of most fluorophores. In this paper, estimates of the anisotropy contributions from flexible and global motions of the IgG-SpA complexes are determined by contrasting theoretical and measured decays. For this analysis it was assumed that each of the experimental phi L-values is a weighted composite of the rotational correlation time associated with the less restricted flexible motions of the Fab arms, phi F, and the correlation time associated with global tumbling of the entire particle, phi G. A general two-exponential expression was used to relate phi F and phi G to phi L. This approach was meaningful because phi G-values of the various SpA complexes had been calculated from hydrodynamic measurements. The theoretical decays clearly show that, even if phi G is much longer than phi F, these two rotational motions still cannot be resolved over the experimentally accessible time range. Families of emission anisotropy decay curves for IgG antibodies with different amounts of intrinsic global coupling and for anchored antibodies with different amounts of steric hindrance were simulated by varying the preexponential weighting factors of the flexible and global terms. By comparing the calculated curves with the measured decays, it is evident that the rabbit IgG anti-dansyl antibodies do not have much intrinsic global coupling, but rather they are highly flexible. The curves also indicate that even for the exceptionally compact IgG4-SpA2 17-S complex, which showed the most steric hindrance in electron micrographs, the appropriate phi G weighting factor is only 0.28. Thus, as supposed earlier, the anchored antibodies exhibit considerable segmental flexibility. In closing, the above concepts are used to examine the results of     

Modulation of IgM to IgG class switch by protein A.

The Fc binding property of soluble protein A (SpA) from Staphylococcus aureus has been utilized to form IgG-SpA complexes which enabled an increase or decrease of IgG from the host depending on the dose of SpA administered. When 5 micrograms SpA was administered the IgG-SpA complexes were rapidly catabolized and, hence, low plasma IgG levels were observed. In contrast 25 micrograms SpA resulted in a significant increase in the IgG level in the host plasma. Based on these observations, the present investigation attempted to study the effect of IgG depletion/increase on the primary and secondary B-cell response to sheep erythrocyte (SRBC) antigen in Balb/c mice. Introduction of 5, 10 and 25 micrograms SpA at the time of the primary antigenic challenge inhibited both the primary IgM and the secondary IgM and IgG responses in a dose-dependent manner. Administration of 5 and 10 micrograms SpA at the time of the secondary antigenic challenge enabled the host to maintain the otherwise depressed secondary IgM response equivalent to the normal primary response. In contrast, 25 micrograms SpA at the time of the secondary antigenic challenge inhibited both the IgM and IgG PFC responses. These results extend our understanding of the mechanism of switch in immunoglobulin class expression during antigen driven maturation of the B-cell response.     

Effect of protein A and its fragment B on the catabolic and Fc receptor sites of IgG

Radiolabeled protein A from Staphylococcus aureus (SpA) injected i.v. into mice and rabbits forms a soluble [(IgG)2-(SpA)1]2 complex (Mr = 684 000) which is identical in composition to that formed by SpA in vitro with an equivalent amount or an excess of IgG. A soluble rabbit IgG-SpA complex injected into a mice or rabbits dissociates completely in vivo and a new complex is formed with the IgG of the recipient animal. The half-life of SpA administered to a mouse or a rabbit is therefore the half-life of the IgG-SpA complex formed in vivo. In mice and rabbits the half-life of the complexes formed is 9 and 30 h, respectively, whereas the half-life of rabbit IgG in these animals is 106 and 153 h, respectively. Fragment B of SpA (fSpA) reacts with IgG of mouse and rabbit and forms an (IgG)1-(fSpA)1 complex. Complexes of identical composition are formed if fSpA is injected i.v. into mice and rabbits. The half-life of the complexes in mice and rabbits are much shorter than those of the corresponding free IgG in these animals (up to 15 times). This result suggests that the binding of fSpA to the CH2 and the CH3 domains of IgG alters the function of the site, which controls the catabolism of IgG and is located in the CH2 domain. By contrast, fSpA does not change the Fc receptor-binding site of IgG, indicating that the Fc receptor site and the catabolic site are unrelated to each other.     

Structure of IgG and IgY molecules in ribosome—Antibody complexes as studied by electron microscopy

The overall shape and dimensions of IgG (rabbit) and IgY (chicken) antibodies against ribosomal proteins have been studied in electron micrographs of ribosome—antibody complexes. The antibodies appear as Y-shaped molecules with an angle of about 90° between their Fab arms. The length of one Fab arm amounts to about 10 nm. No differences between the IgG and IgY molecules could be detected electron microscopically. The data obtained on the shape of IgG and IgY correlate with those of earlier electron microscopic studies while the determined size of the Fab arms is in the range found by scattering methods.     

Extracellular portions of HLA antigens are not compact globulae

Membranes of human spleen cells were hydrolyzed by papain and the extracellular portions of HLA antigen molecules isolated by monoclonal antibodies fixed on Sepharose. The isolated proteins were spin-labeled by TEMPO-dichlorotriazine. The values of rotational correlation times (tau) of spin-labeled proteins were calculated using dependencies of magnetic parameters found from ESR spectra vs viscosity at constant temperature. The tau-values were equal to 8 nsec for class I molecules and 14 nsec for class II molecules. These values were significantly lower than those predicted for a rigid sphere with dimensions equal to the extracellular portions of HLA molecules (20 nsec). This fact suggests the existence of flexibility in poly-functional HLA molecules, which seems to be important for their biological activity. In this respect, extracellular portions of HLA molecules resemble flexible Fc fragments (tau = 12 nsec) and differ from rigid Fab fragments (tau = 20 nsec) of immunoglobulins G. The rotation of the oligosaccharide chains attached to HLA molecules is restricted.     

Simultaneous detection of two different cell surface antigens by electron microscopy by means of multivalent hybrid antibody with double specificity

Multivalent hybrid antibody (MHA) complexes with dual specificity were prepared by combining two antibodies of different specificities, one against ferritin (Fer) the other against horseradish peroxidase (HRP), with protein A of Staphylococcus aaureus (SpA).Electron microscopy of mouse spleen lymphocytes and thymocytes (previously coated with mouse IgG anti-Thy-1 antibody) treated with IgG anti-Fer/SpA/IgG anti-mouse Ig complex and Fer gave better resolution and higher accuracy than previously obtained with IgG anti-HRP/SpA/IgG anti-mouse Ig complex and HRP (Mandache et al., 1980). Surface Thy-1 alloantigen and Fc receptor (charged with human IgG) treated with a mixture of IgG anti-Fer/SpA/IgG anti-Thy-1 and IgG anti-HRP/SpA/IgG anti-human Fab could be simultaneously detected on the thymocyte surface by either light or electron microscopy using Fer and HRP. The concomitant visualization of Thy-1 alloantigen (with Fer) and FcR (with HRP) on mouse thymocyte clearly shows that their distribution is largely independent and that the amount of Thy-1 antigen is greater.These results show that electron microscopy with a mixture of MHA is a useful technique for simultaneous location of two antigenic markers on the cell surface.     

Fc-Dependent Monoclonal IgG1-Mediated Suppression of Antibody Response

It has been known for a long time that passively administered antibodies (Abs) or immune complexes regulate the immune response to their specific antigen (Ag). IgG may sometimes suppress the humoral immune response against soluble antigens. The exact mechanism behind this phenomenon has not been understood yet and the requirement for the Fc part is still a matter of controversy. The present study was undertaken to clarify whether there is a true IgG-mediated Fc-dependent suppression of the immune response. Antigen and monoclonal antibody (mAb) used in this study were recombinant human interferon gamma (r-hIFN-γ) and mouse monoclonal antibodies specific for human IFN-γ [anti-hIFN-γ mAb (CAy-IFNγ38)] respectively. An intact IgG-free preparation of Fab plus various Fc fragments was prepared from papain-digested CAy-IFNγ38. Ag/IgG and Ag/Fab complexes were prepared at various molar ratios. Keeping the Ag doses constant, mice were immunized either with Ag, Ag/IgG or Ag/Fab complexes. Primary immunization and the boosting were performed with the samples in complete and incomplete Freund's adjuvants respectively. Specific antibody levels were measured by an ELISA. Immunization performed with Ag/Fab complexes even at a molar ratio of 1:1.36 did not result in marked suppression of the response when compared to that of Ag only-immunization. In contrast, Ag/IgG complexes resulted in nearly 90% suppression of the antibody response. Our observations suggest that Fc part of IgG molecule plays a crucial role in suppression of the in vivo antibody response against the Ag when complexed with intact IgG.     

S. aureus IgG-binding proteins SpA and Sbi: host specificity and mechanisms of immune complex formation

The evasion of the host immune response is central to the pathogenicity of Staphylococcus aureus, and is facilitated by the ability of the cell wall-associated protein A (SpA) to bind immunoglobulin G Fc fragments, thereby impeding phacocytosis and classical pathway complement fixation. SpA also acts as a B-cell superantigen through interactions with the heavy-chain variable part of Fab fragments, and sequesters immunoglobulins by forming large insoluble immune complexes with human IgG. Here we show that the formation of insoluble immune complexes is mediated by the binding of (VH3+) Fab fragments in addition to Fc, and that SpA forms soluble complexes with IgG Fc fragments. We compared these results with those for Sbi, a second staphylococcal immunoglobulin-binding protein, and note that this protein requires only the Fc fragment for precipitation with human IgG. Homology models of immunoglobulin-binding domains of SpA and Sbi in complex with Fc reveal the molecular basis of the Fab-independent formation of insoluble complexes by Sbi. Finally, we compared the sequences of the spa and sbi genes from human strains to those infecting a range of animal hosts to determine whether Sbi and SpA have acquired specificity for host IgG. We note remarkable sequence conservation within the IgG-binding domains of these genes, consistent with a lack of host specificity. The Fab-independent binding of IgG by Sbi could have significant clinical implications. The use of SpA in immunoadsorption therapy can cause severe side-effects, thought to be mediated by Fc gamma R recognition and complement fixation. The formation of insoluble immune complexes with Sbi occurs only via Fc binding and free Fc regions are unlikely to be available for Fc gamma R recognition and complement fixation.     

Fc‐Dependent Monoclonal IgG1‐Mediated Suppression of Antibody Response

It has been known for a long time that passively administered antibodies (Abs) or immune complexes regulate the immune response to their specific antigen (Ag). IgG may sometimes suppress the humoral immune response against soluble antigens. The exact mechanism behind this phenomenon has not been understood yet and the requirement for the Fc part is still a matter of controversy. The present study was undertaken to clarify whether there is a true IgG‐mediated Fc‐dependent suppression of the immune response. Antigen and monoclonal antibody (mAb) used in this study were recombinant human interferon gamma (r‐hIFN‐γ) and mouse monoclonal antibodies specific for human IFN‐γ [anti‐hIFN‐γ mAb (CAy‐IFNγ38)] respectively. An intact IgG‐free preparation of Fab plus various Fc fragments was prepared from papain‐digested CAy‐IFNγ38. Ag/IgG and Ag/Fab complexes were prepared at various molar ratios. Keeping the Ag doses constant, mice were immunized either with Ag, Ag/IgG or Ag/Fab complexes. Primary immunization and the boosting were performed with the samples in complete and incomplete Freund's adjuvants respectively. Specific antibody levels were measured by an ELISA. Immunization performed with Ag/Fab complexes even at a molar ratio of 1:1.36 did not result in marked suppression of the response when compared to that of Ag only‐immunization. In contrast, Ag/IgG complexes resulted in nearly 90% suppression of the antibody response. Our observations suggest that Fc part of IgG molecule plays a crucial role in suppression of the in vivo antibody response against the Ag when complexed with intact IgG.     

Interspecies hybrid antibody with dual specificity

The preparation of soluble multivalent hybrid antibody by protein A of Staphylococcus aureus (SpA) is limited exclusively to rabbit IgG because other species have SpA-precipitating IgGs. Starting from a soluble complex consisting of one rabbit IgG antibody molecule linked to one SpA molecule (rabbit IgG anti-A/SpA), an interspecies hybrid antibody with dual specificity was prepared using either mouse or human IgG antibody, the molecular formula of this complex being (rabbit IgG anti-A/SpA/mouse or human IgG anti-B)₂. These complexes are useful tools for the investigation of cell surface antigens against which no appropriate antibody can be raised in rabbits.     

Radioimmunoassays for protein A of Staphylococcus aureus

Radioimmunoassays have been developed that can detect nanogram amounts of protein A (SpA), a product generated by Staphylococcus aureus that binds selectively to the Fc region of IgG from most mammalian species. Competition assays for fluid phase SpA utilize antibodies produced in chickens, 125I-labeled SpA as the tracer molecule, and either F(ab')2 fragments of rabbit IgG anti-chicken IgG or 40% ammonium sulfate as the precipitating agent to separate antigen-antibody complexes from free antigen. The double antibody assay could be carried out in serum from species that form only soluble complexes with SpA (e.g., rabbit), that react poorly with SpA (e.g., rat), or under appropriate conditions in serum from species (e.g., dog) that show high reactivity with SpA and form precipitating complexes. Chicken antibodies prepared by affinity chromatography on SpA-Sepharose and labeled with 125I were used in a direct binding assay for SpA present either on the cell wall of Cowan strain I or Wood 46 bacteria, in insoluble complexes prepared from SpA and whole serum or purified IgG, or in Clq binding complexes that were formed by passage of serum from normal or tumor bearing humans or dogs over SpA-collodion charcoal. Since both types of assays could detect SpA even in the presence of serum or IgG, they offer advantages over other techniques in which the SpA-Fc interaction may interfere.     

Purification of antibodies to bacterial antigens by an immunoadsorbent and a method to quantify their reaction with insoluble bacterial targets.

A combination of procedures was employed to develop a radioimmunoassay which quantified the binding of antibodies to antigens of either intact Propionibacterium acnes or to antigens of insoluble extracts derived from the bacteria. Reactive antibody populations were purified by use of bacterial immunoadsorbents which were prepared by coupling P. acnes to diethylaminoethyl cellulose. Binding of antibodies was detected with [125I]staphylococcal protein A ([125I]SpA) and optimal conditions for the assay defined by varying the amounts of antibodies, bacterial antigenic targets and [125I]SpA. In antibody excess, 100% of available [125I]SpA was bound by the target-antibody complexes. However, when antibody concentration was limiting, a linear relationship was demonstrated between per cent specific binding of [125I]SpA and antibodies bound to bacterial targets. These results were achieved only with immunoadsorbent-purified antibody populations and not with hyperimmune sera or IgG. The radioimmunoassay detected subtle antigenic differences and similarities between P. acnes, P. acnes extracts and a variety of unrelated microorganisms.     

Interaction between rat peritoneal macrophages and sensitised erythrocytes: dependence on IgG subclass, antibody density and the degree of hapten conjugation.

The interaction between macrophages (M phi) and antibody sensitised target cells was studied by the use of rat peritoneal macrophages, TNP hapten conjugated sheep red blood cells (SRBC) and homologous antibodies of subclasses IgG1 and IgG2a. Under optimal conditions, the great majority of the M phi formed rosettes with IgG1-sensitised antibodies while a maximum of 50% was achieved when target cells were sensitised by IgG2a. Using a double rosette technique, the major part of rosette-forming cells was found to bind both of the isotypes. IgG1-mediated rosette formation was observed at very low degrees of sensitisation as opposed to IgG2a-mediated target cell binding. Not only the amount of bound antibody but also the degree of hapten conjugation (epitope density) appear to influence the ratio of rosette-forming cells. IgG1-mediated rosette formation was partially inhibited by monomeric IgG1 and more efficiently by soluble ovalbumin (OVA)-anti-OVA complexes involving IgG1-type antibodies, while IgG2a mediated rosette formation was inhibited by OVA-anti-OVA complexes containing IgG2a type antibodies, and less efficiently by complexes involving IgG1. No inhibition was found by monomeric IgG2a. Based on the present data, we propose that two types of receptors are involved in the interaction of M phi and target cells coated by IgG1 and/or IgG2a type antibodies. One requires a multiple antibody-receptor interaction, binding both subclasses at overlapping binding sites; the other is able to interact with IgG1 and does not depend on the multiplicity of interactions.     

Direct demonstration of multiple VH allotopes on rabbit Ig molecules: allotope characteristics and Fab arm rotational flexibility revealed by immunoelectron microscopy

Immune complexes composed of rabbit IgG and Fab fragments of antibody specific for the VH framework allotypes were analyzed by molecular immunoelectron microscopy. In this manner, the number of allotypic epitopes ( allotopes ) and their approximate topological location could be determined. A monoclonal anti-allotype Fab was used to show that relatively fine details of allotope location and orientation are demonstrable with this technique. Each of the three VH allotypes (a1, a2 and a3) was found to express at least two spacially separated allotopes . The locations of the allotopes varied for each of the three allotypes. Some allotopes were observed near the VH-CH1 switch region while others were in close proximity to the complementarity-determining region or were at intermediate positions. In addition to the information concerning allotope topology, the variety of configurations resulting from the interaction of monoclonal antibody Fab with IgG suggest considerable rotational flexibility (twisting) of the Fab arm of IgG about its long axis.     

Conformational changes in the antibody constant domains upon hapten-binding

Bacterial proteins A and G (SpA and SpG) are immunoglobulin receptors that can be used as probes for monitoring change in the conformation of heavy chain constant (C(H)) domains. Interaction of anti-(4-hydroxy-3-nitrophenyl)acetyl (NP) antibody (Ab) with SpA and SpG were measured by isothermal titration calorimetry and surface plasmon resonance in order to address the question of whether hapten-binding induces a conformational change in the C(H) domain. The interactions of IgG2a or its enzymatic fragments with SpA were measured in the presence or absence of the hapten. Although binding of Fab and F(ab')2 fragments were not observed to free SpA, they did bind to immobilized SpA. In addition, the association constant (K(a)) for interaction of IgG2a with immobilized SpA was approximately 20-fold higher than that with free SpA. This was explained in terms of high avidity resulting from multivalent interaction between IgG2a and immobilized SpA on the chip. Interestingly, the hapten-binding weakened the interaction between the F(ab')2 fragment and SpA. Furthermore, approximately half of the IgG2a was incapable of binding to immobilized SpA in the presence of hapten. These results were explained using a model which assumed the formation of two kinds of SpA/IgG complexes; one through sites on F(ab')2 arms and the other through sites on the Fc region. The former type dissociated as a result of hapten-binding, as did the F(ab')2 fragment and suggested that a conformational change had occurred around the Fab arms, while the latter type did not dissociate because of the higher avidity of the Fc region. However, using a mutant SpA with a lower K(a) value for the interaction with IgG2a, it was shown that hapten-binding induced long range conformational changes in the Fc region of IgG2a. Similar evidence of conformational change upon hapten-binding was also obtained using SpG as a probe.     

Electron microscopic and immunochemical analysis of the broadly neutralizing HIV-1-specific, anti-carbohydrate antibody, 2G12

2G12 is one of only a few cloned antibodies with broadly neutralizing specificity to HIV-1 envelope proteins. Crystallographic and electron microscopic (EM) data showed that the Fab arms are locked together via a novel VH domain exchange. Both the conventional and the unprecedented additional VH-VH antigen binding sites show specificity for high mannose oligosaccharides on the silent face of gp120. We have now extended the EM and biochemical analysis of 2G12. Unligated 2G12 IgG1 molecules clearly show paired (parallel attached) Fab arms in the "doughnut" configuration attached to the Fc both in individual and computationally averaged images. A minority of the IgG molecules in the 2G12 prep showed the open "Y" configuration of conventional IgG. The averaged EM image compares well to the atomic structure model of 2G12. Papain digests of 2G12 yielded paired Fab arms (Fab dimer), as observed by EM, which dissociated into Fab-sized fragments in non-reducing SDS-PAGE. Purified 2G12 reduced and alkylated H and L chains can reassociate to form IgG molecules with the Fab dimer configuration and can combine with L and H chains from conventional human IgG to form hybrid molecules. 2G12 is heavily aggregated following brief acid exposure possibly as a result of its unique structure. A model of the aggregation process is proposed. An anti-Id MAb was shown by EM to react with neither the conventional nor additional antigen binding sites, but bound to the lateral faces of the Fab arms of intact, reduced and alkylated, and reconstructed 2G12 molecules. Efforts to identify IgG molecules with a similar intertwined Fab dimer structure in a large IgG pool were unsuccessful.     

Antigenic change in a human IgG4-specific CH3 epitope upon binding of a monoclonal antibody against a neighboring IgG4-specific epitope.

Two sets of monoclonal antibodies (mAbs) probably reacting with two different epitopes in the CH3 domain of the human IgG4 molecule were studied. We observed that the commercially available mAb HP 6011 inhibited the antigen binding of the three mutually inhibitable mAbs, 40-A2, 41-E8 and 43-F11 (40-series), made by us. However, the 40-series mAbs, including those with similar affinity such as mAb HP6011, were not able to inhibit mAb HP 6011. When the 40-series mAbs were preincubated with IgG4, the mAb HP 6011 could partially displace these antibodies. This one-way inhibition indicates that upon binding mAb HP 6011 changes the antigenic structure of the IgG4 molecule by disrupting the epitope for the 40-series mAbs. A steric hindrance of this epitope by mAb HP 6011 is more unlikely, since the small Fab fragment of mAb HP 6011 also inhibited the reaction of the 40-series mAbs.     

The use of staphylococcal protein A in the mixed agglutination test

A mixed agglutination (MA) test employing staphylococcal protein A (SpA) instead of an antiglobulin reagent has been developed for the detection of IgG antibodies bound to cell-surface antigens. In the SpA MA test, a single indicator system may be used for the detection of IgG antibodies from several mammalian species. The sensitivity of this test compares favorably with that of the conventional MA test in the detection of most mammalian IgG. The main advantage of the SpA MA test is its usefulness in the study of antigens on cells which bear Fc receptors. The conventional MA test cannot be conducted with such cells because of adherence of antibody-coated indicator erythrocytes.     

Enhanced antiglobulin-mediated neutralization of herpes simplex virus-IgG complexes by complement and heterologous anti-immunoglobulin

Summary The ability of IgG anti-Fc and anti-Fab to neutralize infectious herpes simplex virus-IgG (HSV-IgG) complexes was determined. When limiting amounts of antiglobulin were used, antibody directed against the Fab portion of human IgG was significantly more effective than anti-Fc antibodies in neutralizing the HSV-IgG complexes. The detection of viral bound antibody was enhanced by the incorporation of heterologous antiglobulin or complement in the antiglobulin neutralization test. Specifically, HSV-IgG which had been incubated with rabbit antihuman globulin was further neutralized by goat antirabbit IgG or guinea pig serum complement. This augmented neutralization test could prove useful in detecting small amounts of antibody bound to virus in infectious isolates from patients or experimental animals with viral diseases.With 4 Figures     

The hemolytic activity of (Fab-Fc) recombinant immunoglobulins with specificity for the sheep red blood cells

Antibody-like molecules were formed by artificial recombination of proteolytic IgG fragments (Fab′ with anti-SRBC activity, and Fc) and used for studies concerning the complement fixation. Such molecules, when composed of single Fab′ bound to Fc fragment appeared inactive, while species containing two Fab′ fragments revealed the hemolytic activity.The results were discussed and interpreted, assuming that the interaction of Fab domains with C_H2 domains in the Fc fragment is a main structural effect influencing the binding of the complement.