Influence of subunit-specific antibodies on the activity of the F0 complex of the ATP synthase of Escherichia coli. II. Effects of subunit c-specific polyclonal antibodies.

After incubation of F1-stripped everted membrane vesicles with antibodies against subunit c of the ATP synthase of Escherichia coli the proton translocation through the open F0 channel was blocked. Rebinding of F1 to those vesicles is affected by the antibody concentration used. In general, the use of F(ab')2 or Fab fragments prepared from anti-c antibodies gave similar results. However, using Fab fragments a higher amount of antigenic binding sites was necessary to block the F0 complex completely, whereas extremely low amounts of Fab fragments were necessary to inhibit the binding of F1. This can be explained by an antigenic determinant of subunit c, which is only accessible to the smaller Fab fragments with a molecular mass of approximately 50,000. Incubation of F1-containing everted membranes with anti-c antibodies showed that the binding of the antibodies resulted in a displacement of F1, while simultaneously the proton translocation through F0 has been blocked. Such a displacement can only be observed after incubation with IgG molecules or F(ab')2 fragments. Fab fragments were not able to displace the F1 part, indicating that the ability of antibodies and F(ab')2 fragments to produce cross-links is responsible for the loss of F1 from the membranes.     

Location of antigenic epitopes on antibody molecules

Using X-ray crystallographic co-ordinates of immunoglobulins, surface regions accessible to a large spherical probe, comparable in size to an antibody domain, were computed. Locations of these exposed regions were compared with those of experimentally determined antigenic sites, i.e. idiotypic, allotypic and isotypic serological markers. In all cases, an excellent agreement was found. The most prominent computed epitopes correspond to convex parts of antibody surface made by reverse turn segments of the polypeptide chain. The computed epitopes occur in homologous positions in all the immunoglobulin domains, and most of the beta-sheet surfaces on the domains are poorly antigenic. The CH2 domain (Fc fragment) has many more antigenic sites than the Fab fragments (antigen-binding fragments). Variable domain epitopes (idiotypes) involve both hypervariable and framework residues, and only about 25% of the hypervariable residues are strongly antigenic. The results indicate that, in a vertebrate body, each antibody molecule may be recognized, and its concentration regulated, by at least 40 complementary anti-immunoglobulin antibodies; therefore, a possibility of an "immune network" with much higher connectivity than is generally assumed should be seriously contemplated.     

Enzymatic and immunochemical properties of lysozyme—Restriction of recognition of lysozyme antigenic determinants in pigs

Pigs immunized with lysozyme responded by producing only nonprecipitating antibody throughout the immunization period. Fig antilysozyme antibodies were found to be resistant to papain fragmentation, only 33% of the antibodies were fragmented with papain. From the binding of fluorescein labeled or ¹⁴C-labeled lysozyme to antilysozyme antibodies it was concluded that the antibodies elicited in pigs recognized only two antigenic determinants of lysozyme. These results were confirmed from the binding of Fab fragments to ¹⁴C-lysozyme. Fab fragments prepared from precipitating rabbit antilysozyme antibody bound ¹⁴C-lysozyme at a molar ratio of Fab/lysozyme = 3. Therefore nonprecipitating antibodies are the outcome of recognition of only two antigenic determinants on lysozyme and inability to form a lattice structure when antibody and antigen interact. This work emphasizes the limitations of using antibodies as a biological reagent for delineating the antigenic determinants on proteins.     

Conformation of PrP(C) on the cell surface as probed by antibodies

We have investigated the conformation of Syrian hamster PrP(C) on the surface of transfected CHO cells by performing cross-competition experiments between a set of nine monoclonal antibody fragments (Fab) directed to defined epitopes throughout the protein. No competition was observed between antibodies recognizing epitopes located within the unstructured N-terminal portion of PrP(C) and those recognizing epitopes located within the ordered C-terminal half of the molecule. However, competition was observed between antibodies recognizing overlapping epitopes and between antibodies recognizing epitopes lying adjacent to one another in the PrP sequence. Titrating the reactivity of each Fab against cell-surface PrP(C) revealed a clear heterogeneity in the accessibility of different specific epitopes. Fab D18, recognizing sequence incorporating the first alpha-helix of PrP(C), bound the largest fraction of the cell-surface PrP population. In contrast, Fab E123, binding an epitope at the extreme N terminus of PrP, and Fab 13A5, binding an epitope in the central region of PrP, were able to recognize fewer than half the number of PrP(C) molecules bound by Fab D18. The pattern of antibody reactivity we observed may, in part, result from N-terminal truncation of a proportion of PrP(C) molecules found at the cell surface. However, truncation cannot account for the marked disparity between exposure of the Fab D18 and 13A5 epitopes, which lie adjacent in the PrP sequence. The relative inaccessibility of the 13A5 epitope likely reflects either PrP(C)-PrP(C) interaction, interaction between PrP(C) and other constituents on the cell membrane, or the existence of PrP(C) subspecies with distinct conformations.     

The recognition domain for the low density lipoprotein cellular receptor is expressed once on each lipoprotein particle

The stoichiometry of binding of monoclonal antibodies and Fab fragments to LDL was assessed. Increasing amounts of two [125I]-labelled antibodies which define epitopes at or near the LDL-receptor recognition domains of apoB were incubated with fixed amounts of LDL and antibody-LDL complexes were separated from free antibodies by heparin-MnCl2 precipitation. Saturation kinetics were obtained and data were analyzed according to Scatchard. One antibody or Fab fragment was bound per LDL particle. Homogeneity of binding was indicated by straight Scatchard lines and by the binding of virtually all LDL particles by an antibody affinity chromatographic column.     

Epitope mapping of pneumococcal surface protein A of strain Rx1 using monoclonal antibodies and molecular structure modelling

Pneumococcal surface protein A (PspA) is an antigenic variable vaccine candidate of Streptococcus pneumoniae. Epitope similarities between PspA from the American vaccine candidate strain Rx1 and Norwegian clinical isolates were studied using PspA specific monoclonal antibodies (mAbs) made against clinical Norwegian strains. Using recombinant PspA/Rx1 fragments and immunoblotting the epitopes for mAbs were mapped to two regions of amino acids, 1-67 and 67-236. The discovered epitopes were visualized by modelling of the PspA:Fab part of mAb in three dimensions. Flow cytometric analysis showed that the epitopes for majority of mAbs were accessible for antibody binding on live pneumococci. Also, the epitopes for majority of the mAbs are widely expressed among clinical Norwegian isolates.     

Interaction of antibodies with liposomes bearing fluorescent haptens

Kinetic and equilibrium aspects of the recognition of antigenic model membranes by antibodies have been studied. Monoclonal anti-fluorescein IgG and its monovalent Fab fragment were allowed to interact with a fluorescein-lipid hapten that was incorporated into phospholipid vesicles. The binding was assayed in the nanomolar hapten concentration range by monitoring the quenching of hapten fluorescence by antibody. The rate and strength of the binding depended on the lipid composition of the vesicles; cholesterol enhanced both. The biphasic binding kinetics observed at high antibody concentrations for some compositions, plus additional spectroscopic evidence, led us to hypothesize that the hapten existed in a composition-dependent equilibrium between at least two conformations: (1) extended away from the membrane surface, available for binding, and (2) sequestered at or in the surface, unavailable for binding. The rate and strength of IgG binding were always greater than those of Fab, indicating bivalent binding by the IgG. This binding was intra-vesicular, since no agglutination of the vesicles was detected.     

The neuraminidase of influenza virus

It is the enzyme neuraminidase, projecting from the surface of influenza virus particles, which allows the virus to leave infected cells and spread in the body. Antibodies which inhibit the enzyme limit the infection, but antigenic variation of the neuraminidase renders it ineffective in a vaccine. This article describes the crystal structure of influenza virus neuraminidase, information about the active site which may lead to development of specific and effective inhibitors of the enzyme, and the structure of epitopes (antigenic determinants) on the neuraminidase. The 3-dimensional structure of the epitopes was obtained by X-ray diffraction methods using crystals of neuraminidase complexed with monoclonal antibody Fab fragments. Escape mutants, selected by growing virus in the presence of monoclonal antibodies to the neuraminidase, possess single amino acid sequence changes. The crystal structure of two mutants showed that the change in structure was restricted to that particular sidechain, but the change in the epitope was sufficient to abolish antibody binding even though it is known in one case that 21 other amino acids on the neuraminidase are in contact with the antibody.     

Monoclonal antibodies against the lac carrier protein from Escherichia coli. 2. Binding studies with membrane vesicles and proteoliposomes reconstituted with purified lac carrier protein

Monoclonal antibodies 4B1 and 5F7 bind to distinct, nonoverlapping epitopes in the lac carrier protein. By use of immunofluorescence microscopy and radiolabeled monoclonal antibodies and Fab fragments, it is shown that both 4B1 and 5F7 bind to spheroplasts and to right-side-out vesicles, but only to a small extent to inside-out vesicles. Clearly, therefore, the lac carrier protein has an asymmetric orientation within the cytoplasmic membrane of Escherichia coli, and both epitopes are located on the periplasmic surface. In right-side-out vesicles, radiolabeled 4B1 binds with a stoichiometry of 1 mol of antibody per 2 mol of lac carrier protein, while radiolabeled 4B1 Fab fragments bind 1:1. Importantly, the intact antibody and its Fab fragments bind to proteoliposomes reconstituted with purified lac carrier protein with a stoichiometry very similar to that observed in right-side-out membrane vesicles. Thus, it seems highly likely that the orientation of the lac carrier protein in the reconstituted system is similar to that in the bacterial cytoplasmic membrane, at least with respect to 4B1 epitope.     

Multiple overlapping epitopes in the three antigenic regions of horse cytochrome c1

To gain a better understanding of the diversity of epitopes on a protein, the specificities of 103 monoclonal antibodies to a model antigen, horse cytochrome c(cyt c), were analyzed. The antibodies were generated in in vitro monoclonal, secondary antibody responses against horse cyt c coupled to hemocyanin in splenic fragment cultures. For this assay, horse cyt c-primed murine B lymphocytes were transferred to irradiated, hemocyanin-primed recipients. A panel of seven mammalian cyts c differing at one to six residues out of 104 and cyanogen bromide-cleaved fragments of horse cyt c containing residues 1-65, 1-80, and 66-104 was used to examine the specificities of the antibodies. Twenty-two distinct reactivity patterns were observed, even though the majority of the monoclonal antibodies were found to bind in the three previously identified antigenic regions of the molecule about residues 44-47, 60-62, and 89-92. The results indicate that each of the three antigenic regions consists of multiple overlapping epitopes. Few of the antibodies directed to any given antigenic region bound polypeptide fragments inclusive of the epitope sequences, demonstrating that some antibodies were more conformationally dependent than others. Only 13% of the antibodies bound to cyanogen bromide-cleaved polypeptide fragments that together encompassed the entire length of the protein. Considering the large number of antibodies analyzed and the reoccurrence of 13 of the 22 clonotypes in different lymphocyte donors, it is likely that the antibody specificities tabulated herein approach yet do not completely enumerate the total inventory of the horse cyt c-specific B cell repertoire. The remarkable diversity for epitope recognition within antigenic regions observed here is likely to pertain to protein antigens in general, and strongly supports the widely held notion that the entire surface of a protein is potentially antigenic. The restriction of the epitopes of horse cyt c to three antigenic regions where the amino acid sequences of the mammalian cyts c differ probably results from tolerance of the mice to their own cyt c.     

Patterns of antibody specificity during the BALB/c immune response to hen eggwhite lysozyme.

We tested 49 BALB/c antilysozyme mAb from seven intervals during the immune response to lysozyme for patterns of specificity and avidity. We found that the antibody epitopes in composite covered at least 80% of the lysozyme surface, and their patterns of overlap suggest a continuum of potential antibody epitopes. Previously observed regional specificities, which emerged at different times in the immune response, were more discretely defined in late response antibodies, when the majority of mAb could be assigned to one of three functionally nonoverlapping complementation groups. The area covered by each antigenic region may be greater than an individual epitope, and may include multiple epitopes that overlap structurally and functionally to varying degrees. Connectivity between antigenic regions was seen in interactions among early and late stage antibodies, and among secondary stage mAb, but not among tertiary stage mAb from hyperimmunized mice. Patterns of overlap of early and late response antibodies suggest that the organization of antibody specificities change during the progression from primary to secondary to tertiary response. Over the same period in the response, the average relative avidity of IgG1 kappa mAb did not increase, suggesting that "affinity maturation" of serum antibodies reflects an increase in the number and diversity of antibodies, rather than an overall increase in the avidity of individual antibodies.     

Electron and X-ray diffraction studies of influenza neuraminidase complexed with monoclonal antibodies

Complexes of influenza virus neuraminidase both with antigen-binding (Fab) fragments and with whole monoclonal antibody molecules have been crystallized. Uniformly thin platelet microcrystals suitable for structure analysis by electron diffraction, yielding reflections to approximately 4.3 A resolution, have been grown from one neuraminidase-Fab complex, that of N9 neuraminidase with 32/3 Fab, and thicker crystals of a second neuraminidase-Fab complex (N9 neuraminidase-NC35 Fab) diffract X-rays to approximately 4.0 A resolution. Electron microscope lattice images of microcrystals both of Fab and of immunoglobulin G complexed with neuraminidase have been interpreted in terms of negatively stained images of the respective individual complex protomers. The sites of binding of the antibodies to the antigen are consistent with the notion that single amino acid changes observed in monoclonal variants of neuraminidase occur in binding epitopes for the antibody used for their selection.     

Loss of rotational mobility of band 3 proteins in human erythrocyte membranes induced by antibodies to glycophorin A.

The effect of antibodies to glycophorin A on the rotational diffusion of band 3 in human erythrocyte membranes was investigated by transient dichrosim. Three antibodies that recognize different epitopes on the exofacial domain of glycophorin A all strongly reduce the rotational mobility of band 3. The effect is at most only weakly dependent on the distance of the epitope from the membrane surface. The degree of immobilization obtained with two of the antibodies, BRIC14 and R18, is very similar to that produced by antibodies to band 3 itself. Similar results were obtained with membranes stripped of skeletal proteins. Fab fragments and an antibody to glycophorin C had no effect on band 3 rotational mobility. These results rule out a mechanism whereby band 3 rotational immobilization results from enhanced interactions with the membrane skeleton that are mediated by a conformational change in glycophorin A. Rather, they strongly indicate that the antibodies to glycophorin A cross-link existing band 3-glycophorin A complexes that have lifetimes that are long compared with the millisecond time scale of the transient dichroism measurements.     

Light chain of natural antibody plays a dominant role in protein antigen binding

Examinations of the contribution and the specificity of heavy (H) and light (L) chains of natural antibodies to antigen binding may help us to better understand antigen recognition and the development of naive B cells. We previously generated natural Fab antibody fragments reactive to preS1 of HBV using a naive, non-immunized Fab antibody library derived from peripheral B cells of a normal healthy volunteer. We now constructed expression vectors for the Fd (VH + CH1), L chain, and scFv fragments using the sequences encoding parental Fabs as a source of natural antibody genes. The recombinant antibody fragments were expressed as inclusion bodies in E. coli BL21 (DE) cells. When denatured and then refolded, the antibody fragments retained their binding properties. Recombinant L chains and scFvs exhibited three- to 40-fold higher affinities (in the order of 10(7) M(-1)) over the parental Fabs, whereas the affinities of Fds (in the order of 10(5) M(-1)) were much lower compared to the parental Fabs. The results obtained from sandwich ELISA revealed that the L chains bound the virus more efficiently than Fds. Additional experiments were performed to evaluate the specificity of the recombinant fragments for surface proteins of HBV. Fds and L chains were reactive towards HBsAg and the preS2 peptide as well as preS1 and showed patterns of epitope recognition quite different from those of parental Fabs. The data presented here demonstrate that the prominence of the L chain in determining protein binding activity is a property of natural antibodies and is quite unlike the antibodies induced by immunization, and that the specificity of Fab is not determined by the individual antibody chain but by the correct pairing of H and L chain.     

Binding of a monoclonal antibody and its Fab fragment to supported phospholipid monolayers measured by total internal reflection fluorescence microscopy.

The association of an anti-dinitrophenyl monoclonal antibody and its Fab fragment with supported phospholipid monolayers composed of a mixture of dipalmitoylphosphatidylcholine and dinitrophenyl-conjugated dipalmitoylphosphatidylethanolamine has been characterized with total internal reflection fluorescence microscopy. The surface densities of bound antibodies were measured as a function of the antibody and Fab solution concentrations, and as a function of the solution concentration of dinitrophenylglycine. The apparent association constant of Fab fragments with surface-associated haptens was approximately 10-fold lower than the association constant for haptens in solution, and the apparent surface association constant for intact antibodies was only approximately 10-fold higher than the constant for Fab fragments. Data analysis with simple theoretical models indicated that, at most antibody surface densities, 50-90% of membrane-associated intact antibodies were attached to the surface by two antigen binding sites.     

Chimeric fab fragments of antibodies to recombinant Ebola virus glycoprotein

Previously, we have determined the nucleotide and amino acid sequences of the variable domains of three mouse monoclonal antibodies specific to the individual epitopes of the Ebola virus glycoprotein: GPE118 (IgG), GPE325 (IgM) and GPE534 (IgG) [1]. In the present paper, chimeric Fab fragments of Fab118, Fab325, and Fab534 antibodies were obtained based on the variable domains of murine antibodies by attaching CH1 and CL constant regions of human kappa-IgG1 to them. The recombinant chimeric Fab fragments were synthesized in the heterologous expression system Escherichia coli, isolated and purified using metal chelate affinity chromatography. The immunochemical properties of the obtained Fab fragments were studied by immunoblotting techniques as well as indirect and competitive ELISA using recombinant Ebola virus proteins: EBOV rGPdTM (recombinant glycoprotein of Ebola hemorrhagic fever virus without the transmembrane domain), NP (nucleoprotein) and VP40 (structural protein). The identity of recombinant chimeric Fab fragments, as well as their specificity to the recombinant glycoprotein of Ebola hemorrhagic fever virus (EBOV GP) was proved. The results of indirect ELISA evidence the absence of immunological cross-reactivity to NP and VP40 proteins of Ebola virus. The dissociation constants of the antigen-antibody complex K _d equal to 5.0, 1.0 and 1.0 nM for Fab118, Fab325 and Fab534, respectively, were determined; they indicate high affinity of the obtained experimental samples to EBOV GP. The epitope specificity of Fab fragments was studied using a panel of commercial neutralizing antibodies. It was found that all studied antibodies to EBOV GP are targeted to different epitopes, while the epitopes of the recombinant chimeric Fab fragments and original murine monoclonal antibodies (mAbs) coincide. All the obtained and studied mAbs to EBOV GP are specific to epitopes that coincide or overlap the epitopes of three commercial neutralizing mAbs to Ebola virus: epitopes Fab118 and Fab325 overlap the epitope of the known commercial mAb h13F6; Fab325 epitope also overlaps mAb c6D8 epitope; Fab534 epitope is located near mAb KZ52 conformational epitope, in the formation of which amino acid residues of GP1 and GP2 domains of EBOV GP are involved.     

Internalization of blocking antibodies against mannose-6-phosphate specific receptors.

Antibodies against mannose-6-phosphate specific receptors inhibit the receptor-dependent endocytosis of exogenous lysosomal enzymes as well as the sorting of endogenous lysosomal enzymes. This inhibition was correlated with an apparent loss of the receptors. We report here that treatment of cells with the antibody results in the formation of receptor-antibody complexes that are not extracted by the procedure used for the solubilization of receptors prior to immunoprecipitation and detection of the receptor. The apparent loss of receptors is observed with both native antibody and the F(ab)2 fragments, but not with Fab fragments. In contrast the transport of lysosomal enzymes is inhibited by all three forms of the antibody. The inhibition is ascribed to masking by the antibody of the enzyme-binding site in the receptor. The inhibition of the sorting of endogenous lysosomal enzymes by antibodies added to the medium indicates that the mannose-6-phosphate specific receptors at the sorting site are in dynamic equilibrium with those at the cell surface. The receptor-antibody complexes formed at the cell surface appear to cycle between the cell surface and intracellular membranes. A fraction of the internalized antibodies dissociates from the receptors and is degraded after transfer into lysosomes. Complexing with Fab increases the concentration of the receptor in the lysosomes and decreases 2- to 3-fold the half-life of the receptor.     

Interaction of rabbit IgG with anti-IgG: localization of epitopes and absence of immunoreactivity of the pFc'-fragment]

To localize essential epitopes of rabbit IgG, a series of proteolytic IgG fragments obtained by papain (Fab, Fc) or pepsin (pFc', F(ab')2) proteolysis have been prepared and their interaction with sheep antibodies against rabbit IgG has been studied. The data obtained suggest that essential immunoreactive epitopes of rabbit IgG are located in the CH2 domain and hinge region. This finding is in line with the results obtained by computing the antigenic sites of immunoglobulins. However, the deviation from the computed antigenic structure was deduced from the complete lack of immunoreactivity of the pFc fragment, it being a dimer of the terminal CH3 domain of the Fc fragment. The hinge region comparable in size with the dimensions of the epitope reveals high affinity binding to anti-IgG, thus testifying to the localization of the expressed epitope or its essential part in the hinge region. Proteolytic cleavage of this region leads to a significant decrease in the binding of the IgG fragment to anti-IgG. In addition to the CH2 domain and hinge region, a relatively low interaction of the antigen-binding antibody fragments with anti-IgG was found.     

Nonlabeled secondary antibodies augment/maintain the binding of primary, specific antibodies to cell membrane antigens.

BACKGROUND: In studies on surface membrane antigen expression using immunofluorescence techniques, it is commonly observed that direct staining gives weaker signals than the signals following indirect staining with fluorochrome-conjugated secondary antibodies. This is most marked when cells have also been permeabilized in order to stain intracellular protein. The commonly accepted explanation for this observation is that fluorochrome-conjugated secondary antibodies bind to a higher number of binding sites on the primary antibody, as compared to the binding of conjugated primary antibodies to the membrane antigens. Another hypothesis might be that the antibody/antibody complexes formed on the membranes when using the indirect technique may have an augmented ability to bind the membrane epitopes. The present study was performed in order to check this hypothesis. MATERIALS AND METHODS: Peripheral blood mononuclear cells were stained with fluorochrome-conjugated anti-CD antibodies directly without or with a second-step application of nonconjugated goat anti-mouse IgG antibodies, followed by different fixation and permeabilization methods. The cells were analyzed by flow cytometry. RESULTS: A second-step application of nonconjugated goat anti-mouse IgG antibodies following direct staining with fluorochrome-conjugated anti-CD antibodies gave a significant increase in membrane antigen expression on permeabilized cells as compared to direct staining alone. The secondary antibody must be bivalent, since whole IgG or F(ab')(2) fragments of the goat anti-mouse antibodies showed effects, while Fab fragments did not. CONCLUSIONS: Nonlabeled secondary antibodies are able to influence the binding of primary, specific antibodies to cell membrane antigens on cells treated with permeabilizing agents necessary for staining intracellular proteins. The improved membrane antigen expression seems to be due to the formation of a network of primary and secondary antibodies on the cell surface, with increased ability for maintaining binding to CD antigens.     

Epitopes of herpes simplex virus type 1 glycoprotein gC are clustered in two distinct antigenic sites.

Epitopes of herpes simplex virus type 1 (HSV-1) strain KOS glycoprotein gC were identified by using a panel of gC-specific, virus-neutralizing monoclonal antibodies and a series of antigenic variants selected for resistance to neutralization with individual members of the antibody panel. Variants that were resistant to neutralization and expressed an antigenically altered form of gC were designated monoclonal antibody-resistant (mar) mutants. mar mutants were isolated at frequencies of 10(-3) to 10(-5), depending on the antibody used for selection. The epitopes on gC were operationally grouped into antigenic sites by evaluating the patterns of neutralization observed when a panel of 22 antibodies was tested against 22 mar mutants. A minimum of nine epitopes was identified by this process. Three epitopes were assigned to one antigenic site (I), and six were clustered in a second complex site (II) composed of three distinct subsites, IIa, IIb, and IIc. The two antigenic sites were shown to reside in physically distinct domains of the glycoprotein, by radioimmunoprecipitation of truncated forms of gC. These polypeptides lacked portions of the carboxy terminus and ranged in size from approximately one-half that of the wild-type molecule to nearly full size. Antibodies recognizing epitopes in site II immunoprecipitated the entire series of truncated polypeptides and thereby demonstrated that site II resided in the N-terminal half of gC. Antibodies reactive with site I, however, did not immunoprecipitate fragments smaller than at least two-thirds the size of the wild-type polypeptide, suggesting that site I was located in the C-terminal portion. Sites I and II were also shown to be spatially separate on the gC polypeptide by competition enzyme-linked immunosorbent assay with monoclonal antibodies representative of different site I and site II epitopes.