Antibody-antigen binding constants determined in solution-phase with the threshold membrane-capture system: binding constants for anti-fluorescein, anti-saxitoxin, and anti-ricin antibodies.

Affinities of various monoclonal and polyclonal antibodies for fluorescein-containing antigens, saxitoxin and ricin, were determined by using a light addressable potentiometric sensor-based system (Threshold). The dissociation constants, determined from Scatchard plots, ranged from 2 x 10(-7) to approximately 3 x 10(-12) M. Dissociation constants for fluorescein and saxitoxin were compared with values determined by independent means. This technique was found to be quick, simple, reproducible, and accurate.

Light-addressable potentiometric sensor for biochemical systems.

Numerous biochemical reactions can be measured potentiometrically through changes in pH, redox potential, or transmembrane potential. An alternating photocurrent through an electrolyte-insulator-semiconductor interface provides a highly sensitive means to measure such potential changes. A spatially selectable photoresponse permits the determination of a multiplicity of chemical events with a single semiconductor device.

Receptors for C3b on the neutrophil surface associate with the cytoskeleton-containing cell residue subsequent to cell lysis with non-ionic detergent. Cytoskeletal-CR1 interactions.

Radioiodinated Fab' and F(ab')2 fragments were prepared from 57F monoclonal antibody which is specific for the C3b receptor (CR1) on human cells. We find that both Fab' and F(ab')2 fragments bind to CR1 on human neutrophils and then dissociate, at 0 degree C, with half-lives of 172 and 35 min, respectively. In addition to binding to cell surface CR1, both antibody fragments bind specifically to the isolated non-ionic detergent-insoluble cellular residue (NDIR) which remains after cell lysis and solubilization in Triton X-100. We also find that Fab' and F(ab')2 antibody fragments remain associated with the NDIR after the radiolabeled antibody fragments are allowed to bind to cell surface CR1, the unbound fragments removed, and the cells subsequently lysed in non-ionic detergent. Because the cytoskeletal matrix (together with the cell nucleus) makes up a substantial portion of the NDIR, these results suggest that some fraction of cell surface CR1 may be associated in vivo with the cytoskeletal matrix. However, post-lysis association of cell surface-CR1-antibody fragment complexes to the NDIR occurs. Examination of the binding kinetics of this interaction reveals that less than 10% of cell surface CR1 exists bound to the NDIR prior to cell lysis. While the nature of post-cell lysis association of CR1 with the NDIR is unknown, several modulating effects are noted. For example, binding of bivalent cross-linking agents to CR1 on cells followed by a 37 degrees C incubation is known to induce internalization of cell surface CR1. We find that binding of F(ab')2 antibody fragments under these conditions causes a 50% decrease in association of this 57F fragment to the NDIR. Pretreatment of the cells at 37 degrees C with the chemotactic peptide N-formyl-methionyl-leucylphenylalanine similarly caused a 50% decrease of (Fab'-labeled) CR1 association with the NDIR cell fraction. These results support the hypothesis that chemotactic peptide serves to enhance CR1-mediated adherence to C3b-bearing targets by increasing the density of CR1 on the cell surface rather than by inducing cytoskeletal-dependent, detergent-stable, CR1 redistribution on the cell surface.

Superoxide enhances photobleaching during cellular immune attack against fluorescent lipid monolayer membranes.

Lipid hapten-containing monolayer membranes with bound, anti-hapten antibody molecules serve as model immunological target membranes. Targets with bound-IgG trigger guinea pig macrophages to (a) adhere, (b) spread, (c) release lysosomal enzymes, and (d) increase cyanide-insensitive oxygen consumption. When the target membranes are derivatized with fluorescein, there is a 2-3-fold enhancement in the rate of fluorescein photobleaching in regions of cell-monolayer contact. This effect is due to release of O2- from macrophages, as shown by inhibition with superoxide dismutase and by the fact that enhanced photobleaching is not observed with cells of the RAW264 macrophage line, which undergo responses (a)-(d), but do not release O2- extracellularly. The O2- dependent photobleaching reaction appears to be relatively specific for fluorescein, as it did not occur with two other fluorophores, 4-nitrobenz-2-oxa-1,3-diazole and tetramethyl-rhodamine. Because stimulated neutrophils release large quantities of O2-, the photobleaching of fluorescein-labeled target membranes in response to neutrophils was examined. Monolayer membranes with specifically bound IgG caused neutrophils to adhere and become markedly motile during incubation at 37 degrees C. Like macrophages, neutrophils induced O2- -dependent photobleaching of fluorescein-labeled IgG in regions of cell-monolayer contact. In addition, neutrophils gave rise to a slower, nonphotochemical loss of fluorescence in the same contact regions. The latter effect is apparently due to cleavage of target-bound fluorescent IgG by proteolytic enzymes secreted by the neutrophils in response to the target surface.

Monoclonal antibodies to a nitroxide lipid hapten.

The isolation and characterization of a hybridoma cell line producing a monoclonal IgG1 antibody against a spin-label nitroxide group is described. The antibody recognizes a synthetic hapten containing linked dinitrophenyl and 2,2,6,6-tetramethylpiperidinyl-1-oxy groups, having an affinity of 3.6 +/- 1.0 . 10(6) M-1 for the soluble hapten at 25 degrees. The antibody binds to phospholipid vesicles containing 2 mol% of spin label-derivitized lipid (lipid hapten) with an affinity of 1.5 +/- 0.2 . 10(8) M-1. This monoclonal IgG1 mediates the binding of hapten-bearing lipid vesicles to mouse macrophage RAW264 cells bearing Fc receptors. The cellular responses to this binding are similar to those observed previously using polyclonal rabbit anti-hapten IgG. As with the heterogeneous antibodies, the monoclonal IgG1 is more efficient in mediating cellular uptake when the vesicles are in the "fluid' physical state (dimyristoylphosphatidylcholine at 37 degrees C) compared to "solid' (dipalmitoylphosphatidylcholine at 37 degrees C). Despite the enhanced binding of "fluid' phospholipid vesicles to cells, only the "solid' vesicles triggered a significant respiratory burst in Raw264 macrophages.

Lipid monolayer-coated solid surfaces do not perturb the lateral motion and distribution of C3b receptors on neutrophils.

We have used epifluorescence and photobleaching techniques to study the lateral distribution and motion of fluorescein-conjugated Fab fragments of anti-C3b receptor antibody bound to human neutrophils when the cells rest on various solid supports (microscope slides or cover slips). Supports composed of quartz, glass, or alkylated glass induced cellular adhesion, spreading, and an extensive lateral redistribution of C3b receptors (but not HLA antigens). The neutrophil C3b receptors become patchy, and the patches apparently undergo nonrandom translational motion. Many patches are found on the upper surfaces of the cells removed from the region of cell membrane-glass contact. In contrast, neutrophils supported by lipid monolayer-coated glass do not adhere or spread, and the C3b receptor remains uniform and diffuses freely (D approximately equal to 2 X 10(-10) cm2/s).

Neutrophil activation monitored by flow cytometry: stimulation by phorbol diester is an all-or-none event.

The population dynamics of single-cell stimulation was analyzed by monitoring autofluorescence by flow cytometry. Stimulation of the respiratory burst in human neutrophils by 12-O-tetradecanoyl phorbol-13-acetate (TPA) caused a decline in highly fluorescent cells (characteristic of resting neutrophils) and a corresponding increase in the number of weakly fluorescent cells (characteristic of activated neutrophils). Increasing concentrations of TPA caused increasing numbers of cells to shift from the highly fluorescent population to the weakly fluorescent population without the appearance of intermediate populations. Thus the neutrophil respiratory burst, a component of neutrophil cytotoxic response, is triggered in an all-or none fashion.

Specific antibody-dependent interactions between macrophages and lipid haptens in planar lipid monolayers.

The binding of guinea pig peritoneal macrophages to planar lipid monolayers on alkylated glass is shown to be highly specific, requiring lipid hapten in the monolayer and specific anti-hapten IgG antibody. This is true for both "fluid" and "solid" monolayer membranes, in which the lateral diffusion coefficients of fluorescent lipid probes and bound fluorescent antibodies differ by at least two orders of magnitude. The region of the (macrophage membrane)--(monolayer membrane) contact is readily observed by using an epifluorescence microscope and fluoresceinated IgG antibodies or fluorescent lipids. The fluorescence intensity of IgG antibodies bound to lipid haptens in fluid monolayer membranes in the region of the (macrophage membrane)--(monolayer membrane) contact was significantly enhanced in the early stages of binding (first 10 min at 24 degrees C), due to a diffusive flux of the fluorescent antibodies into the region of membrane--membrane contact. Cellular activation takes place immediately during a 10-min warm-up to 37 degrees C and can be recognized by rapid symmetrical cell spreading, the formation of "black holes" around the cells (probably due to superoxide-facilitated photochemical bleaching of the fluorophore), and the release of the lysozomal enzyme cathepsin B. Specific antibody-dependent [1-14C]-glucose oxidation by these macrophages on fluid and solid monolayers is quite similar to that reported previously for fluid and solid bilayer vesicle target membranes. These results are significant for understanding the molecular interactions between membranes that are necessary for a macrophage cytotoxic response.

Disappearance of macrophage surface folds after antibody-dependent phagocytosis.

We have employed the method of Burwen and Satir (J. Cell Biol., 1977, 74:690) to measure the disappearance of surface folds from resident guinea pig peritoneal macrophages after antibody-dependent phagocytosis. Unilamellar phospholipid vesicles containing dimyristoylphosphatidylcholine and 1 mol % dinitrophenyl-epsilon-aminocaproyl-phosphatidylethanolamine, a lipid that possesses a hapten headgroup, were prepared by an ether injection technique. These vesicles were taken up by macrophages in a time- and temperature-dependent fashion. Vesicles that contained ferritin trapped in the internal aqueous volume were identified within macrophages by transmission electron microscopy. Scanning electron microscopy has shown that macrophage surface folds decrease dramatically after phagocytosis. The surface fold length (micrometer) per unit smooth sphere surface area (micrometer2) decreases from 1.3 +/- 0.3 micrometer-1 to 0.53 +/- 0.25 micrometer-1 when cells are incubated in the presence of specific anti-DNP antibody and vesicles at 37 degrees C. No significant effect was observed in the presence of antibody only or vesicles only. Our studies shown that phagocytosis is associated with a loss of cell surface folds and a loss of cell surface area, which is consonant with current views of the endocytic process. On the basis of our uptake data, we estimate that approximately 400 micrometer2 of vesicle surface membrane is internalized. The guinea pig macrophage plasma membrane has a total area of approximately 400 micrometer2 in control studies, whereas the cells have roughly 300 micrometer2 after phagocytosis. These estimates of surface areas include membrane ruffles and changes directly related to changes in cell volume. We suggest that during antibody-dependent phagocytosis a membrane reservoir is made available to the cell surface.

Specific antibody-dependent phagocytosis of lipid vesicles by RAW264 macrophages results in the loss of cell surface Fc but not C3b receptor activity.

We have observed that the specific antibody-dependent phagocytosis of lipid vesicles (containing 1 to 2% phospholipid hapten) results in a loss of Fc surface receptor activity from RAW264 macrophages. The Fc-receptor activity was measured by the ability of the macrophages to form rosettes with antibody-coated sheep erythrocytes (E). C3b-receptor activity was monitored similarly by measuring rosetting of IgM and complement-coated sheep erythrocytes (EAC). Under the same conditions, there was no loss of C3b-receptor activity. This suggests that Fc but not C3b cell surface receptors are depleted during specific IgG-mediated phagocytosis. The rosette assay for receptors, in accord with a previous kinetic analysis of vesicle uptake, showed that the depletion of Fc receptors after phagocytosis of specific antibody-coated vesicles was highly dependent on the number of vesicles taken up but relatively independent of the density of IgG bound to the vesicle surface. In contrast to the above results the nonspecific phagocytosis of latex beads results in the loss of both Fc- and C3b-receptor activity as measured by rosetting experiments.

Triggering of the macrophage and neutrophil respiratory burst by antibody bound to a spin-label phospholipid hapten in model lipid bilayer membranes.

The specific antibody-dependent stimulation of the respiratory burst (cyanide-insensitive oxygen consumption, 1-C-glucose oxidation) of RAW264 macrophage cell line by haptenated lipid vesicles depends strongly on the physical properties of the lipid membrane, as well as the surface density of antibodies on the vesicles. Lipid membranes that are "solid" at 37 degrees C (dipalmitoylphosphatidylcholine, DPPC) are much more effective, per vesicle bound, than are "fluid" membranes (dimyristoylphosphatidylcholine, DMPC). Vesicle membranes that have both fluid and solid regions (DPPC containing < 20 mol % cholesterol) show both enhanced binding rates (due to the fluid regions) and enhanced respiratory rates (due to the solid regions). In contrast to these results, the specific antibody-dependent respiratory burst of neutrophils due to haptenated vesicles parallels the antibody-dependent vesicle binding and shows no significant difference between fluid and solid target membranes.

Polymorphonuclear leukocyte-mediated, antibody-dependent, cellular cytotoxicity against tumor cells: dependence on oxygen and the respiratory burst.

Experiments were done to determine 1) whether the respiratory burst of superoxide anion (O2-) production in polymorphonuclear leukocytes (PMN) is triggered during antibody-dependent killing of tumor cells and 2) whether O2- production is essential for cytotoxicity. Three parameters of the respiratory burst (1-14C-glucose oxidation, oxygen consumption, and O2- release) were increased 2.5- to 7.3-fold during killing of antibody-primed tumor cells by human PMN. Added catalase and superoxide dismutase did not inhibit lysis, possibly because these enzymes were unable to diffuse into the inter-plasma-membrane space between killer and target cells. Evidence for an O2- requirement for cytotoxicity was the fact that concentrations of amobarbital or phenylbutazone sufficient to inhibit the cyanide-insensitive respiration of PMN also inhibited cytotoxicity. Also, hypoxic conditions inhibited cytotoxicity from 29 to 73%. The requirement for oxygen was most likely related to O2- generation and not mitochondrial respiration since cyanide and azide, which inhibit mitochondrial respiration, increased cytotoxicity.

Protection against carbon tetrachloride-induced lipid peroxidation in the rat by dietary vitamin E, selenium, and methionine as measured by ethane evolution.

Dietary vitamin E, selenium (Se), and methionine were tested for their ability to inhibit carbon tetrachloride (CCL4)-induced lipid peroxidation. Peroxidation, in vivo, was monitored by the evolution of ethane, an autoxidation product of omega-3-unsaturated fatty acids. Weanling rats were fed a basal diet low in vitamin E, Se, and sulfur-containing amino acids, or diets individually supplemented with these factors. After 3 to 7 weeks, the rats were injected with CCL4 (ip) and ethane was collected for 9 hours. Cumulative ethane evolution was increased by CCl4 in all groups. Vitamin E, Se, and methionine reduced ethane evolution from CCl4-treated rats by 82%, 74%, and 60%, respectively. The toxicity of CCl4 was decreased in correlation with ethane evolution. Thus, methionine and Se, probably by maintaining intracellular glutathione and glutathione peroxidase, protected against CCl4-induced lipid peroxidation, as did vitamin E. Substitution of cod liver oil, which is rich in omega-3-unsaturated fat, for lard in the basal diet increased CCl4-induced ethane evolution six-fold. Relative inhibition by the dietary supplements was not changed. Thus, the feeding of cod liver oil greatly increased ethane production which facilitated the detection and measurement of lipid peroxidation in vivo.

Lipid peroxidation in vivo during vitamin E and selenium deficiency in the rat as monitored by ethane evolution.

Ethane evolution was monitored from vitamin E and selenium (Se)-deficient rats to determine if lipid peroxidation occurs in vivo when these rats develop fatal organ lesions. Weanling rats were fed a vitamin E and Se-deficient, or supplemented, diet for 40 to 90 days. Each was then prefasted for 4 hours and fasting was continued for 24 to 40 hours while ethane was collected. Approximately 50% of the doubly-deficient rats died as a result of fasting. Pathological signs included hematuria, lung hemorrhage, and liver necrosis. Ethane evolution increased exponentially 10 to 20 hours before death and then declined 2 hours before death. Rats that survived (at least 5 days after ethane collection) evolved 7.4+/-1.3 nmoles ethane/100 g body weight/24 hours compared to 100+/-6 for rats that died. Supplementation of the basal diet with vitamin E (200 IU/kg), Se (0.2 ppm, as Na2SeO3), or both, completely prevented mortality and reduced ethane evolution values to 0.4+/-0.2, 3.1+/-0.4, or 0.2+/-0.2, respectively. These experiments indicate that lipid peroxidation occurs in vivo as a result of vitamin E and Se deficiency, and the peroxidation process greatly accelerates during the terminal phase of the fatal disease.