New immunolatex spheres: visual markers of antigens on lymphocytes for scanning electron microscopy

Immunomicrospheres: reagents for cell labeling and separation

Immunomicrospheres are specially designed microscopic particles that have antibodies or similar molecules chemically bound to their surfaces. The antibody-coated microspheres react in a highly specific way with target cells, viruses, or other antigenic agents. Immunomicropheres may be synthesized so that they incorporate compounds that are highly radioactive, intensely fluorescent, magnetic, electron opaque, highly colored, or pharmacologically active. These various types of microspheres may be coated with pure, highly specific monoclonal antibodies obtained by the new hybridoma cell cloning techniques or with conventional antibody preparations. Some of the many present and potential applications for these new reagents are (i) new types of radioimmune or immunofluorescent assays, (ii) improved fluorescence microscopy, (iii) separation of cells on the basis of the fluorescent, electrophoretic, or magnetic properties of bound immunomicrospheres, (iv) markers for use in several types of electron or standard light microscopy, and (v) delivery of lethal compounds to specific undesirable living cells. The combination of the various new types of synthetic microspheres and the newly available homogeneous antibodies offers new opportunities in research, diagnosis, and therapy.

Electrophoretic cell separation by means of microspheres

The electrophoretic mobility of fixed human erythrocytes immunologically labeled with poly(vinylpyridine) or poly(glutaraldehyde) microspheres was reduced by approximately 40%. This observation was utilized in preparative scale electrophoretic separations of fixed human and turkey erythrocytes, the mobilities of which under normal physiological conditions do not differ sufficiently to allow their separation by continuous flow electrophoresis. We suggest that resolution in the electrophoretic separation of cell subpopulations, currently limited by finite and often overlapping mobility distributions, may be significantly enhanced by immunospecific labeling of target populations using microspheres.

Fluorescent colloidal gold: a cytochemical marker for fluorescent and electron microscopy

The gold method was further developed for fluorescent microscopy. Gold granules (12 nm in size) were labelled with rhodamine conjugates of Concanavalin A and avidin. The fluorescent markers were used to mark cell wall mannan on the yeast Saccharomyces cerevisiae either by the one-step, or by the two-step method via a biotinyl derivative of ConA. By fluorescence or transmission electron microscopy, the two-step method was found to achieve a higher density of marking.

Polyglutaraldehyde: a new reagent for coupling proteins to microspheres and for labeling cell-surface receptions. II. Simplified labeling method by means of non-magnetic and magnetic polyglutaraldehyde microspheres

Procedures were developed for the synthesis of a new immunoreagent in form of polyglutaraldehyde (PGL) microspheres in sizes ranging from about 50 nm to 1.5 micron. Addition of fluorochromes during synthesis yielded microspheres of high fluorescence intensity. By carrying out the polymerization of glutaraldehyde in presence of iron oxide, magnetic PGL microspheres were produced. Antibody conjugates obtained by interaction of PGL microspheres with immunoglobulins were used to label human red blood cells (RBC) and lymphocytes. A simple method for the separation of magnetically labeled human RBC from unlabeled cells was demonstrated.

Antigen-specific identification and cloning of hybridomas with a fluorescence-activated cell sorter

Myeloma-spleen cell hybrids (hybridomas) producing antibody to mouse immunoglobulin allotypes have been labeled with fluorescent microspheres coupled with myeloma protein antigens. The ratio of specific to nonspecific microsphere binding by viable hybridoma cells was about 100:1. By using a modified fluorescence-activated cell sorter (FACS), selected hybridoma cells in a mixture have been sorted individually into media in microculture wells, where, with thymocyte feeder cells, they developed into clones producing a desired monoclonal antibody. Viable cells were selected by measurement of their light scattering and autofluorescence properties. Rare antibody-producing clones were obtained without laborious screening and repeated subculturing. This technique should expand the range of monoclonal antibodies readily obtained from hybridomas and greatly facilitate the process of obtaining desired hybridomas.

Differences in the redistribution of concanavalin A and wheat germ agglutinin binding sites on mouse neuroblastoma cells

Concanavalin A (Con A), wheat germ agglutinin (WGA), and Ricinus communis agglutinin (RCA) bound with either 125I, fluorescent dyes, or fluorescent polymeric microspheres were used to quantitate and visualize the distribution of lectin binding sites on mouse neuroblastoma cells. As viewed by fluorescent light and scanning electron microscopy, over 10(7) binding sites for Con A, WGA, and RCA appeared to be distributed randomly over the surface of differentiated and undifferentiated cells. An energy-dependent redistribution of labeled sites into a central spot occurred when the cells were labeled with a saturating dose of fluorescent lectin and maintained at 37 degrees C for 60 min. Reversible labeling using appropriate saccharide inhibitors indicated that the labeled sites had undergone endocytosis by the cell. A difference in the mode of redistribution of WGA or RCA and Con A binding sites was observed in double labeling experiments. When less than 10% of the WGA or RCA lectin binding sites were labeled, only these labeled sites appeared to be removed from the cell surface. In contrast, when less than 10% of the Con A sites were labeled, both labeled and unlabeled Con A binding sites were removed from the cell surface. Cytochalasin B uncoupled the coordinate redistribution of labeled and unlabeled Con A sites, suggesting the involvement of microfilaments. Finally, double labeling experiments employing fluorescein-tagged Con A and rhodamine-tagged WGA indicate that most Con A and WGA binding sites reside on different membrane components and redistribute independenty of each other.

Surface morphology of bone marrow lymphocytes. I. Scanning electron microscopy of small lymphocytes bone marrow and spleen

Cell separation techniques and scanning electron microscopy (SEM) were used to characterize the surface morphology of small lymphocytes in mouse bone marrow. Lymphocyte-rich fractions and unfractionated suspensions of bone marrow and spleen cells from 9--10-week-old C3H male mice were glutaraldehyde-fixed, syringed onto gelatin-coated silver membranes, dehydrated in ethanol, infiltrated with amyl acetate, critical point dried, coated with gold-palladium and examined by SEM. High proportions of cells were retained on the membranes. Purified spleen small lymphocytes showed unimodal distribution curves for cell diameter (mode, 3.4 micrometer) and for number of surface microvilli (mode, 55--60). Bone marrow small lymphocytes were identified initially in lymphocyte-rich marrow fractions and in erythroblast-depleted marrow from polycythemic mice as well as in normal whole marrow. The cells resembled spleen small lymphocytes in size distribution and they showed microvilli. However, the number of visible microvilli was lower on small lymphocytes in the bone marrow (mode, 35--40) than in the spleen. While in each small lymphocyte population the total number of microvilli was greater on larger cells than on smaller ones, the density of microvilli per unit area of cell surface tended to decrease with increasing cell size. The results establish that the small lymphocytes in mouse bone marrow, mainly locally-produced immature cells, have villous surfaces, but the number of microvilli per unit cell surface area is less than that on peripheral small lymphocytes, as seen in the spleen. Neither in the bone marrow nor in the spleen are subpopulations of small lymphocytes distinguishable solely by numbers of microvilli. The findings suggest that microvilli on bone marrow small lymphocytes may undergo further development during post-mitotic maturation, surface receptor expression and migration of the cells to peripheral lymphoid tissues.

Islet-cell-surface antibodies in juvenile diabetes mellitus

Using an indirect immunofluorescence test on suspensions of viable, insulin-producing islet cells from rats, we found that 32 per cent (28/88) of insulin-treated patients with juvenile diabetes have islet-cell-surface antibodies in their circulation. These antibodies also occurred in four of nine children with glucose intolerance, in one of 24 healthy children and in nondiabetic children with thyroid disorders. In the diabetic children, the immunofluorescent reaction was inhibited by preadsorption of serum to islet cells but was little affected by preadsorption to rat hepatocytes or erythrocytes or to acetone powders of various rat tissues, including pancreas. These results show that organ-specific, nonspecies-specific antibodies reactive with the cell surface of the islet cells can be present in serum from diabetic children, and provide an approach to investigation of immunopathological aspects of diabetes mellitus.

Murine macrophage-lymphocyte interactions: scanning electron microscopic study

Light and scanning electron microscopic observations revealed murine macrophage-lymphocyte interactions involving the initial contact of peritoneal, spleen, or thymus lymphocytes with peritoneal macrophage processes or microprocesses followed by clustering of lymphocytes over the central nuclear area of the macrophages. Lymphocyte-lymphocyte clustering was not observed in the absence of macrophages. Attachment and subsequent clustering appeared not to require the presence of serum or antigen; the attachment of allogeneic or xenogeneic lymphocytes was comparable to that seen in the syngeneic system, but central clustering of these lymphocytes failed to occur. No attachment or clustering was observed when thymic lymphocytes were cultured with thymus derived fibroblasts rather than with peritoneal macrophages. Lymphocyte attachment to immune, antigen-activated, syngeneic macrophages occurred more rapidly than that to normal unstimulated syngeneic macrophages; however, lymphocytes attached to the "activated" macrophages appeared to be killed by a nonphagocytic mechanism. A similar increase in the rate of lymphocyte attachment to macrophages occurred in the presence of migration inhibitory factor. Subsequent lymphocyte clustering on macrophages was observed in the migration inhibitory factor-stimulated cultures. In addition, lymphocyte-macrophage interactions similar to those in vitro were observed to occur in vivo on intraperitoneally implanted cover slips.

The preparation of latex particles with covalently bound polyamines, IgG and measles agglutinins and their use in visual agglutination tests

Carboxylated latex particles were substituted with side arms terminating in primary amine and hydrazine groups. The particles were coupled to aldehyde groups generated on glycoproteins which were treated with sodium periodate. Particles having the alipathic primary amine putrescine hapten as the sole substituent and particles linked to glycoproteins such as measles agglutinins and IgG were used to detect the presence of the corresponding antibodies or antigens in biological fluids by agglutination tests.

Scanning electron microscopy and the surface morphology of human lymphocytes

The surface of the human lymphocyte as seen by the scanning electron microscope shows variations which may reflect the functional state or environment of the cell. Preparative techniques and future developments in this area are reviewed.

Surface immunoglobulin of human lymphocytes

Using ferritin as surface marker, the localization of the surface immunoglobulin (Ig) was studied on peripheral lymphocytes from normal human individuals and patients with macroglobulinaemia Waldenström by scanning immunoelectron microscopy. Normal IgG-, IgM-lymphocytes and pathological IgM-lymphocytes were then compared with regard to their topographical differences. In all cells examined, IgG- and IgM-conjugated ferritin particles were detected all over the cell surface, but the distribution of the former on the normal IgG-lymphocytes was slightly more diffuse than that of the latter on the normal and pathological IgM-lymphocytes. Furthermore, in the pathological IgM-lymphocytes, the clustered IgM-conjugated ferritin particles were found in great number on the microvilli. Normal IgG-lymphocytes were almost always characterized by short rod-like microvilli standing densely and vertically on the cell surface. Some of normal IgM-lymphocytes had a similar appearance to those of normal IgG-lymphocytes (type A) but others (type B) had tilted rod-like microvilli or wide plate-like processes on their surface. As for IgM-lymphocytes of macroglobulinaemia, most lymphocytes had tilted rodlike mirovilli and wide plate-like processes similar to type B, whereas a minor population of the pathological lymphocytes carried long, thin rod-like microvilli standing vertically on the surface.

Scanning-autoradiography--a new method for the demonstration of membrane-surface-associated structures

A method for the demonstration of membrane-localized markers and receptors is described which results from the combination of autoradiography and scanning electronmicroscopy.

Detection and separation of lymphocytes with specific surface receptors, by using microparticles

Horse anti-(human lymphocyte) globulin was immobilized together with fluorescein labelled dextran in spherical microparticles of polyacrylamide (AHLG-particles). The particles had a diameter of 1-5 micrometer and a density of 1.12g/cm3, with globulin exposed on the surface. Human lymphocytes bearing the antigen (thymus-derived lymphocytes) bound the particles, which were easily detected by fluorescence microscopy. In this way, about 58% of circulating human lymphocytes were able to bind AHLG-particles at 23 degrees C. Non-specific binding was low, only 3% when human serum albumin was present in the buffer, and only 4% when non-specific horse globulins were incorporated in the microparticles. The cell-particle complexes could be separated from cells that had not reacted by density-gradient centrifugation in Ficoll/metrizoate. The viability was not changed after the separation procedure. The number of cells binding AHLG-particles corresponded well the the relative amount of T-cells. When the cells binding AHLG-particles were separated from the lymphocytes, the number of T-cells decreased remarkably, indicating that the antibodies bind preferably to the T-cell population. Concanavalin A immobilized in microparticles was sufficiently exposed to initiate the agglutination of the lymphocytes. The agglutination was completely inhibited by preincubating the microparticles with alpha-methyl mannoside.

An avidin-biotinyl-propranolol complex for beta-adrenergic receptor characterization

The synthesis of biotinyl-hexaglycyl-NEDA (abbreviation:BGN), a biotinyl derivative of propranolol, is described. This bifunctional molecule binds with high affinity to the biotin-binding protein, avidin. The duck erythrocyte was used as a model beta-receptor system. Formation of an avidin-BGN-beta-receptor complex was demonstrated in intact erythrocytes, in erythrocyte ghosts, and in the digitonin-solubilized beta-receptor. The avidin-BGN complex will be used for localization and purification of the beta-receptor.

Immunolatex localization by scanning electron microscopy of intestinal bacteria from cockroaches

An immunolatex reagent was developed from antiserum to Fusobacterium varium, an obligate anaerobe isolated from the colon of the cockroach, Periplaneta americana L. A new technique that enabled the preparation of a highly efficient immunolatex conjugate was used to localize the bacterium with scanning electron microscopy, in situ, in the mixed gut contents.

Human B lymphocytes in giemsa stained preparations

A mixed antiglobulin procedure has been developed to provide simultaneous study of cell surface immunoglobulin and detailed cytological features in Giemsa stained preparations. The percentage of surface immunoglobulin bearing lymphocytes, monocytes and granulocytes changed dramatically when lymphocytes were incubated at 37 degrees C, or following sheep erythrocyte rosette enrichment procedures.

Reactions of human platelets with microspheres of poly(hydroxymethyl methacrylate) and polyacrylamide

We have studied the contact interaction of platelets with hydrogels. In the form of microspheres, 0.6-1.0 mu, poly(glycol methacrylate) (polyHEMA) and poly(methyl methacrylate) beads cause platelets to aggregate at concentrations of about 10(8) beads/ml. Polyacrylamide and (20/80) poly(acrylamide-HEMA) copolymer were ineffective in aggregating platelets. The admixture of 20% methacrylate to polyHEMA rendered the beads inactive. Blood plasma components other than fibrinogen were found essential to the interaction of the beads with platelets. Near-infrared spectra of the hydrogels polyacrylamide and polyHEMA showed the water hydrogen bonds to be the same for both and different from those in pure water. The monomer HEMA is an inhibitor of platelet aggregation and the release reaction at levels of 0.1%. It is concluded that the two hydrogels have different blood compatibilities, which depend more on the network structures than the water structures in the respective gels.

Concanavalin A and wheat germ agglutinin receptors on Dictyostelium discoideum. Their visualization by scanning electron microscopy with microspheres

The cellular slime mold, Dictyostelium discoideum, is a convenient model for studying cellular interactions during development. Evidence that specific cell surface components are involved in cellular interactions during its development has been obtained by Gerisch and co-workers (1, 2) using immunological techniques. Smart and Hynes (3) have shown that a cell surface protein can be iodinated on cells in aggregation phase, but not in vegetative phase, by the lactoperoxidase procedure. Recently, McMahon et al. (4), and Hoffman and McMahon have demonstrated, by SDS gel electrophoresis, considerable differences in cell surface proteins and glycoproteins of plasma membranes isolated from cells at different stages of development. Plant lectins have also been used to monitor changes in cell surface properties of D. discoideum cells during development. Weeks and co-workers (5, 6) have detected differences in the binding and agglutination of cells by concanavalin A (Con A). Gillette and Filosa (7) have shown that Con A inhibits cell aggregation and prematurely induces cyclic AMP phosphodiesterase. Capping of Con A receptors has also been reported (8). Reitherman et al. (9) have recently reported that agglutination of cells by several plant lectins and the slime mold agglutination, discoidin, changes during development. Such studies indicate that differences in surface properties exist for cells at various stages of development. However, owing to the uncertainties in the factors which contribute to lectin-induced cell agglutination (10), the molecular basis for these observations remain to be determined. In this study, we have used microspheres (11-14) coupled to either Con A or wheat germ agglutinin (WGA) as visual markers to study by scanning electron microscopy the topographical distribution of lectin receptors on D. discoideum cells fixed at different stages of development. We also describe the effect of labeling on the distribution of lectin receptors and on the morphology of the cell surface.

The scanning electron microscopy of normal human peripheral blood lymphocytes

A study of the effects of various sample preparation techniques for scanning electron microscopy has been undertaken in an attempt to resolve conflicting descriptions of the surface topography of human peripheral blood lymphocytes. By fixing cells in suspension--a technique thought most likely to avoid the production of artefacts--no clearly defined morphological classification of lymphocytes could be made, and when T- and B-lymphocyte enriched preparations were studied their surfaces appeared similar. Both T- and B-rosetted cells showed identical morphological changes as a result of their interaction with red blood cells. The smooth cells described in other reports were found only under certain conditions of preparation. It is therefore not possible to distinguish between T- and B-cell populations, using the S.E.M., on the basis of surface morphology alone.

Immunospecific labeling of mouse lymphocytes in the scanning electron microscope

Bone marrow-derived (B) and thymus-derived (T) Balb/c mouse lymphocytes were identified in the scanning electron microscope (SEM) by the immunospecific attachment of one of several kinds of large-molecular-weight markers distinguishable in SEM. These markers (tobacco mosaic virus, keyhole limpet hemocyanin, bushy stunt virus, and bacteriophage T4) could be modified with hapten groups and linked with anti-hapten antibody, in an indirect (sandwich) scheme, to hapten-modified anti-cell-surface antibody bound to the cell surface. Hapten-modified antibodies to B cell antigens (goat anti-mouse-immunoglobulin) or to T cell antigens (rabbit anti-mouse brain) were employed to identify these two lymphoid cell types in unfractionated spleen, mesenteric lymph node, bone marrow, and thymus cell populations. The topography of B cells was always indistinguishable from that of T cells. No surface features were found to be unique to either cell type. In suspension, the majority of B and T cells had one or no microvilli regardless of the tissue source of the labeled cells. Cells in suspension that had microvilli (usually 10% of the total cell population) were always unlabeled. However, after cell contact with a glass surface, approximately half of both the B and T cell population had a villous topography.

A scanning electron microscope study of concanavalin A receptors on retinal rod cells labeled with latex microspheres

Con A-methacrylate microsphere conjugates prepared by a two-step glutaraldehyde reaction were used to label Con A-binding sites on bovine rod photoreceptor cells for visualization by scanning electron microscopy. A dense distribution of markers was observed on the surface of the rod outer segment, the inner segment, and the synaptic region. Disk membranes also appear to be heavily labeled with the Con A-microsphere conjugates. The Con A inhibitor, alpha-methyl mannoside, inhibited the binding of the conjugate to the surface of these visual cells.

A new method of in situ hybridization

A new method for gene mapping at the chromosome level using in situ hybridization and scanning electron microscopy is described and has been applied to mapping the rRNA genes of Drosophila melanogaster. Biotin is covalently attached to Drosophila rRNA via a cytochrome c bridge at a ratio of one cytochrome-biotin per 130 nucleotides by a chemical procedure. Polymethacrylate spheres with a diameter of ca. 60 nm are prepared by emulsion polymerization and are covalently attached to the protein avidin at a ratio of 5-20 avidins per sphere. The biotin-labeled rRNA is hybridized to denatured DNA in a chromosome squash. Upon incubation with a sphere solution, some of the biotin sites become labeled with spheres because of the strong non-covalent interaction between biotin and avidin. The chromosome squash is examined in the scanning electron microscope (SEM). Polymer spheres, which are visible in the SEM, are observed to label the nucleolus, where the rRNA genes are located.

Application of latex microspheres in the isolation of plasma membranes. Affinity density perturbation of erythrocyte membranes

Immunolatex spheres, originally developed as visual markers for scanning electron microscopy, were employed as membrane density perturbation reagents. Methacrylate spheres were bound to antibody molecules and used to label antigens on erythrocytes. Ghosts prepared from labeled cells were subjected to isopycnic centrifugation on continuous sucrose and dextran gradients. It was found that the labeled erythrocyte membranes had a substantially higher density than unlabeled membranes. The extent to which the membrane density was shifted on a given gradient depended on the number, size and density of the latex spheres and could be closely predicted by theory. These results suggest that the reagents and techniques described here have potential application for the isolation of plasma membranes from more complex cell types.